University of Southern California Dissertations and Theses

A population genomics approach to the study of speciation in flowering columbines

(USC Thesis Other)

A population genomics approach to the study of speciation in flowering columbines

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content A POPULATION GENOMICS APPROACH TO THE STUDY OF SPECIATION IN
FLOWERING COLUMBINES
by
Elizabeth Cooper
A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(MOLECULAR BIOLOGY)
May 2011
Copyright 2011 Elizabeth Cooper
Dedication
To my family
and especially to my grandfather, John Parker, who was probably the one
person who would have enjoyed reading this thesis in its entirety
ii
Acknowledgments
I would like to thank my advisor, Magnus Nordborg, for the guidance, patience and
support (and office furniture) that he has given me during my time spent at USC. I
would also like to acknowledge Scott Hodges, who has been like a second advisor to me
and without whom none of this project would have been possible. Justin Borevitz has
also been an important part of the Aquilegia project, and I am especially appreciative
of his efforts to bring together everyone working on this system. In addition, I want to
thank the members of my committee: Steve Finkel, Sergey Nuzhdin, Xuelin Wu, and
Dave Conti, who have all provided valuable advice both scientific and otherwise.
I’m grateful towards many of the people that I’ve worked with at USC. Tina Hu was
thefirstpersonImetonmyfirstdayinthelab, andfromdayonewasaconfidanteanda
friend. Chris Toomajian and Alex Platt were excellent teachers to me, and always made
timetoanswermyquestions. HonggangZheng,ChunlaoTang,andMar` ıa-Jos´ eAranzana
taught me everything I know about DNA sequencing. Rana Goyal helped me create a
sequencedatabase. MattiasJakobssonshowedmehowtorunSTRUCTURE,andhelped
me to understand some of my more convoluted results. Bjarni Vilhjalmsson and Daniel
Campo Falgueras have been my 2 very understanding office-mates, and both let me
keep the thermostat at 80
◦
F without the slightest complaint. Other members of the
Nordborg and Nuzhdin labs, both past and present, who have taught me about science
and made my time as a graduate student more enjoyable are Badri Padhukasahasram,
KeyanZhao,SungKim,PeteCalabrese,PeiZhang,DazheMeng,YuHuang,SuziAtwell,
iii
Marie-Stanislas Remigereau, Glenda Willems, Muhammed Ali Amer, Chitiksha Shah,
Maren Friesen, Brad Foley, and Vincent Plagnol.
I’m also indebted to a handful of people outside of USC: Justen Whittall originally
designed the primers and collected many of the samples used for the original sequencing
study;JeremySchmutzattheJGIprovidedmewiththeunpublishedDNAsequencedata
fromtheAquilegia coerulea genomeproject;ChristosNoutsoscreatedaBLASTdatabase
and other helpful online resources for Aquilegia; Quan Long and other members of the
Nordborg lab at the GMI performed almost all of the Solexa sequencing; Nathan Derieg
prepared the DNA samples used in the Solexa sequencing, and Pablo Cingolani went out
of his way to alter his annotation program to work with the non-public Aquilegia files.
Last but not least, I’d like to thank Lauren Holden, my roommate, carpool buddy,
and good friend, who has provided many years of emotional support and understanding.
iv
Table of Contents
Dedication ii
Acknowledgments iii
List of Tables vii
List of Figures viii
Nomenclature xi
Abstract xiii
Chapter 1: Introduction 1
1.1 Aquilegia as a Model System for the Study of Speciation . . . . . . . . . . 4
1.1.1 Overview of Chapters . . . . . . . . . . . . . . . . . . . . . . . . . 6
Chapter 2: Population Genetics in Aquilegia 7
2.1 Levels of Polymorphism and Shared Polymorphism . . . . . . . . . . . . . 8
2.2 Population Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
2.3 Isolation by Distance in the Aquilegia genus . . . . . . . . . . . . . . . . . 13
2.4 Linkage Disequilibrium and Recombination Rate . . . . . . . . . . . . . . 17
2.5 Materials and Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
2.5.1 Sample collection and preparation . . . . . . . . . . . . . . . . . . 19
2.5.2 Fragment amplification and sequencing . . . . . . . . . . . . . . . 20
2.5.3 Sequence alignment and editing . . . . . . . . . . . . . . . . . . . . 26
2.5.4 Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
2.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
Chapter 3: Sympatric vs. Allopatric Speciation 30
3.1 The Isolation-Migration Model . . . . . . . . . . . . . . . . . . . . . . . . 32
3.2 A Simple Model with No Migration . . . . . . . . . . . . . . . . . . . . . . 33
3.3 Migration Rate Estimates . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
3.4 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
v
Chapter 4: Whole Genome Scan for Speciation Genes 37
4.1 Alignment with Aquilegia coerulea . . . . . . . . . . . . . . . . . . . . . . 38
4.2 Genome-wide Levels of Polymorphism and Divergence . . . . . . . . . . . 39
4.3 Coverage and its Confounding Effects . . . . . . . . . . . . . . . . . . . . 41
4.4 Candidate Speciation SNPs . . . . . . . . . . . . . . . . . . . . . . . . . . 44
4.4.1 Anthocyanin Pathway . . . . . . . . . . . . . . . . . . . . . . . . . 47
4.4.2 Inversions and Speciation . . . . . . . . . . . . . . . . . . . . . . . 48
4.5 Materials and Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
4.5.1 Sample Collection, Preparation, and Sequencing . . . . . . . . . . 49
4.5.2 Read alignment and SNP calling . . . . . . . . . . . . . . . . . . . 51
4.5.3 Sliding Window Analyses . . . . . . . . . . . . . . . . . . . . . . . 52
4.5.4 G–Test for Independence . . . . . . . . . . . . . . . . . . . . . . . 52
4.5.5 Annotation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
4.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
Chapter 5: Conclusions 54
Bibliography 56
vi
List of Tables
2.1 Polymorphism Counts for Each Fragment. For each count in bold,
the number of sites represents the number of SNPs plus the number of
indelstreatedassingleSNPs. Numbersinparenthesesrepresentthenum-
berofsiteswithaMinorAlleleFrequency(MAF)>5%and>10%,respec-
tively. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.2 Levels of polymorphism for synonymous and nonsynonymous
sites. Note that LFY is not listed in this table because it did not contain
any nonsynonymous sites. . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.3 Estimates of recombination rate for each fragment.. . . . . . . . . 13
2.4 Summary of Aquilegia samples used in this study. . . . . . . . . . 23
2.5 Primer pairs used to amplify the 9 nuclear loci. . . . . . . . . . . . 25
2.6 Positions of introns, exons, and UTRs in Each Locus . . . . . . . 26
4.1 Reads Mapping in Each Sample. Expected coverage values were
calculated by dividing the number of base pairs in the Aquilegia genome
(301 million) by the total number of mapped base pairs. . . . . . . . . . . 39
4.2 Highly Differentiated Non–Synonymous SNPs. . . . . . . . . . . . 45
4.3 Samples Used for Solexa Sequencing . . . . . . . . . . . . . . . . . . 50
vii
List of Figures
1.1 Differences in Floral Morphology for (A) A. formosa and (B)
A. pubescens. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
2.1 Variation among levels of polymorphism for each species. The
sequences of each of the 9 fragments were grouped according to species,
and θ
W
and Π were estimated separately for each group. Grey bars rep-
resent estimates for A. formosa, and black bars represent A. pubescens. . 10
2.2 ComparisonofminorallelecountsinA. formosa andA.pubescens.
ThehorizontallinerepresentsthemeanallelecountinA. pubescens,while
theverticallinerepresentsthemeanallelecountin A. formosa. Pointsize
reflects the number of comparisons at that point. Species-specific poly-
morphisms correspond to the points along either the very bottom or the
far left of the plot. All other sites correspond to a shared polymorphism.
There are no fixed differences. The average minor allele frequency for any
species-specific polymorphism was 0.105, while the average frequency for
any shared allele was 0.424. . . . . . . . . . . . . . . . . . . . . . . . . . . 12
2.3 STRUCTURE cannot cluster individuals according to species.
Each individual is indicated by a thin line, where the two colors represent
theestimatedmembershipcoefficientsforthe2clusters. Theclusteredness
score for this plot was estimated as 0.26. . . . . . . . . . . . . . . . . . . . 14
2.4 Inferred population structure for 80 Aquilegia individuals. The
results from STRUCTURE are plotted for K=11, which had an average
clusteredness score of ≈ 0.52. Each individual is represented by a thin
horizontal line, with corresponding population and species information
given on either side. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.5 Probability of Different K Estimates: The estimated log probability
ofthedata(ascalculatedbySTRUCTURE)isplottedagainstdifferentK
values. For each K value, STRUCTURE was run 3 times, and the plotted
value is the average of those 3 runs. . . . . . . . . . . . . . . . . . . . . . 16
viii
2.6 Average Clusteredness for Different K Values. For each K value,
the average clusteredness measures the extent to which each individual
belongs to a single cluster rather than to multiple clusters, so the higher
the clusteredness the “better” the clusters.. . . . . . . . . . . . . . . . . . 16
2.7 Phylogeny for Two genomic sequences. . . . . . . . . . . . . . . . . 18
2.8 True Phylogeny for Several Members of the Aquilegia genus.
Figure adapted from the website of S. Hodges . . . . . . . . . . . . . . . . 19
2.9 Relationship between geographic distance and genetic distance.
Each dot represents a comparison between 2 populations of at least 5
individuals. For populations where there were more than 5 individuals,
estimates of F
ST
were bootstrapped to ensure that the larger sample size
did not cause any bias in the estimate. . . . . . . . . . . . . . . . . . . . . 20
2.10 Other factors influencing F
ST
in Aquilegia. In all panels, red dots
indicate comparisons where both populations were the same for the factor
being considered, while gray dots indicate comparisons where the two
populations were different. Panel (A) shows F
ST
vs distance both within
and between species, with the green diamonds indicating comparisons
between either A. formosa or A. pubescens and one of the natural hybrid
populations. Panel(B)showsF
ST
vsdistancewiththesameanddifferent
pollinator syndrome, while Panel (C) shows the same comparisons for
habitat type. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
2.11 Relationship between geographic distance and genetic distance
for A. formosa and A.pubescens only. Black squares represent com-
parisonsbetweenA. formosa andA. pubescens populations;graytriangles
are comparisons among populations of A. formosa; white circles are com-
parisons among populations of A. pubescens. . . . . . . . . . . . . . . . . 22
2.12 R
2
versus distance for the combined data. Only SNPs with a minor
allele frequency≥10% were usedin pairwise comparisons. Trend line was
fitted as described in the Methods section. . . . . . . . . . . . . . . . . . . 22
3.1 The isolation with migration (IM) model with 6 parameters.
Adapted from [23]. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
3.2 MIMAR estimates of the time since divergence between A. for-
mosa and A. pubescens. Both panels show the posterior distributions
ofthetimesincedivergenceincoalescenttimeunits. Inthefirstpanel,the
distribution was generated by running MIMAR with the migration rate
fixed at 0; in the second panel, the prior distribution for the migration
rate was U[0.135,7.39]. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
ix
4.1 Read coverage per base pair for each species. The proportion of
base pairs is calculated using the total number of non-missing bases in
the reference genome (no Ns). The last point on the x–axis represents
the combined data for read coverage ≥176. Red bars are values for A.
formosa and blue bars are values for A. pubescens. . . . . . . . . . . . . . 40
4.2 Coverage correlation between the species. Values for read coverage
= 0 and coverage = 1 are not shown because the proportions are much
higher than those on the rest of the plot. The proportions for these values
are 4% and 1%, respectively. . . . . . . . . . . . . . . . . . . . . . . . . . 40
4.3 Sliding window analysis of polymorphism based on 10–kb win-
dows with no overlap. Grey lines indicated shared polymorphism, red
lines are A. formosa polymorphism, and blue lines are A. pubescens poly-
morphism. Green ticks along the bottom of the graphs indicate positions
of fixed or nearly fixed differences. . . . . . . . . . . . . . . . . . . . . . . 42
4.4 Proportion of High Frequency Differences vs. Within Species
Coverage. Ateachwithin-speciescoveragevalue,theproportionofSNPs
with an allele frequency difference (AFD)≥90% was calculated out of the
total number of SNPs with that coverage. The red dot represents 30X
coverage, which was determined to be the best cut-off value. . . . . . . . . 43
4.5 Allele Frequency Difference Threshold vs. Minimum Cover-
age. Minimum coverage values range between 1 and 100X. The Allele
Frequency Difference (AFD) thresholds range between 5% and 100%, in
increments of 5%. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
4.6 Spatial Pattern of Allele Frequency Differences. The p–value of
each SNP site was calculated based on the observed distribution of AFD
in the whole sample. Different colors within each panel indicate separate
scaffolds, which have been grouped together randomly; positions to the
right of the first scaffold in a panel have been adjusted so that scaffolds
couldbeplottedsidebyside. Thehorizontaldottedlinemarksthe−log
10
p
for an allele frequency difference of 93.5%. . . . . . . . . . . . . . . . . . . 46
4.7 Gene Annotation for Different Subsets of SNPs. . . . . . . . . . . 47
4.8 TheBranchofthePhenylpropanoidBiosyntheticpathwayyield-
ing anthocyanins. Arrows indicate which genes are regulated by Myb
transcription factors in different species. Figure adapted from [19]. . . . . 48
4.9 Map of Sampling Locations. Red points represent A. formosa popu-
lations and yellow points represent A. pubescens. . . . . . . . . . . . . . . 50
x
Nomenclature
F
ST
The Fixation index; a measure of popula-
tion differentiation (genetic distance) based
on genetic polymorphism data. It is a special
case of F-statistics, the concept developed in
the 1920s by Sewall Wright., 11
ρ the population recombination rate, 17
θ the population mutation rate, 8
adaptive radiation the evolution of ecological and phenotypic
diversity within a rapidly multiplying lineage
[58], 3
AFD allele frequency differentiation, 5
allopatric a term referring to organisms whose ranges
are entirely separate such that they do not
occur in any one place together., 6
IM isolation–migration, 6
xi
introgression the movement of a gene (gene flow) from
one species into the gene pool of another by
repeated backcrossing an interspecific hybrid
with one of its parent species., 4
LD Linkage Disequilibrium; the nonrandom asso-
ciation of alleles at two or more loci., 7
parapatric a term from referring to organisms whose
ranges do not significantly overlap but are
immediatelyadjacenttoeachother; theyonly
occur together in the narrow contact zone, if
at all., 5
sympatric a term referring to organisms whose ranges
overlap or are even identical, so that they
occur together at least in some places., 5
xii
Abstract
Aquilegia formosa and Aquilegia pubescens are two closely–related species belonging to
the columbine genus. Despite their morphological and ecological differences, previous
studies have revealed a large degree of intercompatibility as well as little sequence diver-
gence between these two taxa [25, 81], and the genetic mechanisms underpinning repro-
ductive isolation remain unknown. In order to assess the feasibility of a full genome scan
for speciation genes, inter– and intraspecific patterns of variation were compared for 9
nuclear loci; it was concluded that the two species were practically indistinguishable at
the level of DNA sequence polymorphism, indicating either very recent speciation or
continued gene flow. As a comparison, the variation at two loci was analyzed across 30
other Aquilegia species, revealing slightly more differentiation among taxa and evidence
for isolation by geographic distance (which was not the case on a more local geographic
scale).
The extremely low levels of genetic variation found between A. formosa and A.
pubescens at neutral loci was deemed ideal for a genome-wide scan for allele frequency
differences, so this was done using Solexa deep sequencing of pooled samples from each
species. Polymorphisms were identified and annotated based on alignment with the A.
coerulea reference genome, and SNPs with extreme values of allele frequency differentia-
tion (≥93.5%) were selected as candidate speciation genes. Two of these extreme SNPs
caused amino acid changes in the MYB and UGT proteins, both of which are known
components and regulators of the anthocyanin (pigmentation) pathway.
xiii
Chapter 1
Introduction
A naturalist, reflecting on the mutual affinities of organic beings, on their
embryologicalrelations,theirgeographicaldistribution,geologicalsuccession,
and other such facts, might come to the conclusion that each species had
not been independently created, but had descended ... from other species.
Nevertheless,suchaconclusion,evenifwellfounded,wouldbeunsatisfactory,
until it could be shown how the innumerable species inhabiting this world
have been modified ...
Charles Darwin, On the Origin of Species, 1859 [11]
In his introduction to the Origin of Species, Darwin aptly captured a fundamental
question in evolutionary biology–one which is still under investigation today: how do
evolutionary forces create and sustain the incredible amount of diversity seen in species
all over the world? The theory of natural selection is now widely accepted, and the
mechanisms of genetic mutation upon which selection acts (which were unknown to
Darwin)arenowwellunderstood,butthequestionofhownewspeciesarisehasremained
hotlycontested. DespiteDarwin’sinitialemphasisontheimportanceofselection,forthe
majority of the last century, the prevailing viewpoint has been that new species evolve
onlyaftergeographicisolation,withnaturalselectionplayingmoreofasupportingrolein
the process [7]. But as genome sequencing has become increasingly accessible, more and
more studies have been finding that organisms with clear morphological and behavioral
distinctionscanstillsharelargeamountsoftheirgenomes(reviewedin[76]). Asaresult,
there has been a resurgence of interest in the role of natural selection during speciation.
The question of how new species evolve is intrinsically linked to two other crucial
and widely debated questions: What is a species? and Why are there species? While
defining a species may at first seem to be a simple problem, it is in fact the oldest and
probably most difficult question to answer in the field of speciation biology. Dating
1
back to the philosophers Plato and Aristotle, there were discussions about the different
“types” of organisms [33]. While it can easily be said that all species represent a type,
the reverse is not true. All types do not represent a species; some types can be higher
orders of classification, such as “plants” and “animals,” or “mammals” and “reptiles.”
While most of these higher categories have a set list of characteristics that define them,
the lower orders of classification tend to become more difficult to distinguish. So, when
does a type get defined as a species?
Ironically, even though Darwin entitled his famous book the Origin of Species, he
actually states within it that he doesn’t believe a “species” is a real thing, but instead
a theoretical construct used to arbitrarily describe sets of individuals closely resembling
oneanother[11]. Inhisview,species,sub-species,andlocallyadaptedpopulationsallfell
into the general category of “variety,” and were not necessarily different things. From
this perspective, the process of speciation is more or less equivalent to the process of
adaptation.
Although many people agree with Darwin, many others do not, and over the years
numerous“speciesconcepts”havebeenproposedinordertoclarifywhatmakesaspecies
aspeciesandnotsomeothertype. Themostwell-knownspeciesconceptistheBiological
Species Concept (BSC), which is centered around the idea of reproduction and repro-
ductive isolation. Although this concept has existed in some form since the 1600s, it
was best defined by and is generally associated with Ernst Mayr, who said: “Species are
groups of actually or potentially interbreeding natural populations, which are reproduc-
tively isolated from other such groups [45].” Even though the BSC provides the most
commonly accepted definition of a species, it has several flaws. Firstly, it can only be
applied to sexually reproducing organisms, which excludes all bacteria and other uni-
cellular organisms, as well as many plant species. In other words, this species concept
cannot be applied to the majority of living things on Earth. Secondly, there is some
ambiguity as to the extent of reproductive isolation that is required for two groups to
be called separate species. Some would argue only completely non-interbreeding groups
2
can truly be called species, yet there are many plants which have been deemed separate
species based on morphological characters that can still form hybrid zones. Lastly, for
groups that are not “potentially interbreeding” because they live too far apart, it is
unclear whether or not they can be defined as separate species based on just the criteria
put forth in the BSC.
Another concept of speciation which expands a bit on Mayr’s view to include more
organisms is known as the Evolutionary Species Concept (EvSC). It was first introduced
by George Simpson, a paleontologist who was primarily interested in fossil record data,
and it focused on the stability of certain morphological traits over time [61]. A slightly
modified and more modern version of the EvSC states that “the processes of evolution,
including genetic drift, migration, and adaptation, cause there to be a thing, an entity
made up of organisms evolving in concert and that collectively form a species [71].”
Considering a species to be an evolutionary unit does not discount the BSC: without
some degree of reproductive isolation it would not be possible for evolution to act inde-
pendently on two populations. At the same time, the EvSC does not entirely rely on
the presence or absence of sexual reproduction, so it can be applied to more organisms.
Even though there are many other speciation concepts in existence, the EvSC is the one
that most accurately describes the species of columbine discussed in this thesis, and it
also provides a framework for generating and answering questions about how speciation
occurs from a genetics perspective.
Ifaspeciescanbeconsideredasagroupofindividualsuponwhichvariousevolution-
ary forces are collectively acting, then it follows naturally to question what the role and
relative importance of each force is during the process of speciation. Finding the genes
involved in reproductive isolation and analyzing the patterns of genomic variation both
within and outside of these genes can help to reveal the signatures of natural selection
andgeneticdrift, andultimatelyleadtowardsansweringthequestionofhownewspecies
evolve.
3
1.1 Aquilegia as a Model System for the Study of Specia-
tion
Part of Darwin’s evidence and inspiration for the theory of natural selection stemmed
from his observations of beak variation in 14 species of Gal´ apagos finches [10]. Because
the birds were very similar in all aspects except for their beaks, he hypothesized that
“onespecieshadbeentakenandmodifiedfordifferentends”[10]. Itwaslatershownthat
differences in beak morphology allowed the different birds to be adapted to specific food
sources[16]. Thisscenarioofonespeciesdivergingrapidlyintomultiplespeciesoccupying
different ecological niches is known as an adaptive radiation, and such radiations are
invaluable to the study of speciation.
The columbine genus Aquilegia [Ranunculaceae] is an excellent example of a recent,
rapid adaptive radiation [25], and thus should provide an opportunity to identify the
genetic changes important for speciation. The genus is comprised of approximately 70
outcrossing species that occupy a wide variety of habitats in North America, Europe,
and Asia [49] and that differ substantially in floral morphology [49, 80]. Despite these
differences, species are usually cross-compatible [53, 69].
Two species, Aquilegia formosa and A. pubescens, have long been studied for the
purpose of understanding the factors controlling reproductive isolation between them
[21, 4, 26, 17]. A. formosa is found throughout mountainous regions of western North
America while A. pubescens is restricted to the southern Sierra Nevada range [26]. The
species exhibit distinct differences in floral characters that have been shown to influence
pollinator preference, thereby restricting gene flow between them [26, 17](Figure 1.1).
Three of the most crucial floral characters are flower color, nectar spur length, and
flower orientation. Hummingbirds and bees both prefer the red, pendant A. formosa,
while hawkmoths are the primary pollinators of the white, upright A. pubescens. Addi-
tionally, they prefer different habitats: A. formosa populations typically occur in moist
areas with well-developed soils at lower elevations (below 3,000m), whereas A. pubescens
4
populationsarefoundindrier,poorlydevelopedsoilsathigherelevations(3,000-4,000m)
[21,4,27]. However,thetwospeciesarehighlyinterfertile,andformnaturalhybridzones
at mid elevations where the two habitats co-occur [26]. Molecular markers exhibit more
introgression than morphological characters near these zones, suggesting that gene flow
could be extensive between these species for neutral markers [26].
PreviousstudieshaveuncoveredlimitedDNAsequencevariationbetweenA. formosa
and A. pubescens in both chloroplast and nuclear sequences [25, 81]. However, these
previous studies showed either low sequence variation across a wide range of Aquilegia
species [25] or few individuals were sampled [81] and therefore do not address the degree
of genetic differentiation between these species. Other studies suggest that intraspecific
sequence variation may be quite similar in A. formosa and A. pubescens and thus that
theymaybeespeciallyusefulforidentifyingspeciationgenes. Forinstance,microsatellite
locihavesimilarnumbersofallelesandsizeranges[84],andanotherstudyincludingover
850AFLPmarkerspolymorphicinasmallsampleofbothspeciesfoundonlyonemarker
that showed complete differentiation [80].
The purpose of this research is to find the genes responsible for species differences
and assess the evolutionary forces shaping them in order to gain a better insight into
the process of adaptive radiation in this system. The work was done in two stages:
first, a small sequencing survey of inter- and intraspecific polymorphism was done in
order to gauge background levels of differentiation, population structure, recombination,
and mutation. The results from this stage not only laid the foundation for the larger
sequencing study, but also produced a preliminary model for a sympatric or parapatric
speciation scenario. The second stage of the project comprised a high-throughput whole
genomecomparativesequenceanalysisofthetwospecies. Allelefrequencydifferentiation
(AFD) was used as the main criterion for selecting candidate speciation loci.
5
Figure 1.1: Differences in Floral Morphology for (A) A. formosa and (B) A.
pubescens.
1.1.1 Overview of Chapters
In Chapter 2, standard population genetics parameters, such as polymorphism levels,
linkage disequilibrium, F
ST
(differentiation), and population structure were estimated
using nine short nuclear genome sequences in multiple populations of both A. formosa
and A. pubescens. The resultant low levels of differentiation that were found were also
compared to a broader sample of over 30 Aquilegia species.
In Chapter 3, polymorphism data from the sequences in Chapter 2 were used to
simultaneously attempt to address the question of whether incipient speciation with
no gene flow (allopatry), or less recent speciation with ongoing gene flow (sympatry)
would best explain the high levels of shared variation. Estimates of divergence time
and migration rates between species were simultaneously calculated using the Isolation–
Migration (IM) Model [23] framework.
InChapter4, high-throughputIlluminasequencesweregeneratedforpooledsamples
of A. formosa and A. pubescens. ThegenomesweresubsequentlyalignedandSNPswith
extreme values of allele frequency difference were culled from the total sample. These
SNPs were annotated and nonsynonymous changes were identified. Full genomewide
scans of shared and private polymorphism levels were also performed.
6
Chapter 2
Population Genetics in Aquilegia
Mapping and identifying the genes responsible for reproductive isolation is an essential
step in determining the mechanism of speciation in any organism, but it is not the only
step. The evolutionary forces driving divergence are complex and typically involve the
interaction of natural selection, genetic drift, hybridization and gene flow. When species
areinsympatry,thehomogenizingeffectsofgeneflowareexpectedtobecounteractedby
divergent selection on those traits that contribute to reproductive isolation [8, 76, 7], so
the resultant patterns of genetic variation will not be uniform across the whole genome.
However, inherent variation in genomic mutation rates and recombination rates will
also lead to non-uniform patterns of genetic variation, so it is necessary to consider the
genome as a whole in order to provide a context for making inferences about natural
selection.
A knowledge ofpopulationstructure and genome architecture is alsovital for design-
ing any kind of study that will scan for speciation loci. Traditional mapping methods
find genetic markers that are statistically associated with the trait of interest; these
markers are usually not the causal mutation itself, but instead are expected to be in
linkage disequilibrium (LD) with it. Linkage disequilibrium is defined as the nonrandom
association of alleles at two or more loci [16], and it can be influenced by a number of
factors beyond just physical linkage. The extent and variation of LD in any genome
should dictate the number of markers required for a mapping study: if large numbers
of markers tend to be in LD with one another, then any one of them would capture
the variation associated with the trait, and the study would require a low number of
markers in general. Conversely, an overall lack of LD would mean that a very dense set
of markers would be required in order to be assured of finding a marker associated with
7
a particular phenotype. At the same time, once a marker is found, it becomes preferable
tohavelessLD,becausethecandidateregionaroundthemarkerwillbesmaller, making
it easier to find the causal mutation.
Population structure can influence the association between loci and also the associa-
tionbetweenalocusandatrait, soitmustalsobeaccountedforinanyspeciationstudy.
For instance, if A. formosa and A.pubescens populations were highly structured, then
any differences found by comparing only one population from each species might be just
as likely to be the result of random genetic drift as of divergent selection. So, in order
to get an accurate picture of species differences, multiple populations from each species
wouldneedtobe compared, suchthatinterspecific variationcouldbedistinguishedfrom
intraspecific variation.
In order to address these issues and attempt to characterize the neutral genetic
variation in Aquilegia, nine random genomic loci were chosen and sequenced in many
individuals from several populations of both A. formosa and A. pubescens. The results
ofthe analyses performedonthese sequences have beenpreviously published [6], andare
also discussed in detail in the following sections of this chapter.
2.1 Levels of Polymorphism and Shared Polymorphism
The counts of segregating sites found in each fragment are given in Table 2.1. The
population mutation rate (θ) was estimated using both Watterson’s estimator (θ
W
) [78]
and nucleotide diversity (Π) [68], with both estimates generally falling in the range of
0.004 to 0.006 per base pair (Figure 2.1). Overall, these estimates are slightly lower than
estimates ofθ in other outcrossing plant species such as maize (θ≈ 0.0096) [73] and sun-
flowers (θ≈ 0.0094) [35], similar to estimates in the model species Arabidopsis thaliana
[51], and higher than estimates found in soybeans (θ ≈ 0.00097) [86]. However, values
of θ
W
and Π varied across the 9 fragments, and one of them, UF3GT, was substantially
8
more polymorphic than the others, with a value ofθ between 0.01 and 0.02 (Figure 2.1).
Table 2.1: Polymorphism Counts for Each Fragment. For each count in bold, the
number of sites represents the number of SNPs plus the number of indels treated as
single SNPs. Numbers in parentheses represent the number of sites with a Minor Allele
Frequency (MAF) >5% and >10%, respectively.
Fragment Total S
n
Indels A. formosa A. pubescens Shared Fixed
Exclusive Exclusive Diff.
Acetyl 5 (2, 2) 0 1 (0, 0) 2 (0, 0) 2 (2, 2) 0
DEFEN 20 (8, 6) 3 9 (1,0) 4 (0, 0) 7 (7, 6) 0
Gapc 44 (14, 8) 14 18 (1,0) 11 (0,0) 15 (13, 8) 0
H3 18 (7, 2) 2 5 (0, 0) 5 (1, 0) 8 (6, 2) 0
Heat 20 (7, 4) 1 7 (0, 0) 9 (4, 1) 4 (3, 3) 0
AP3 29 (16, 12) 10 10 (4, 3) 11 (4, 2) 8 (8, 7) 0
LFY 12 (4, 2) 4 6 (2, 0) 3 (0, 0) 3 (2, 2) 0
Pist 15 (8, 4) 2 10 (4, 1) 2 (1, 0) 3 (3, 3) 0
UF3GT 34 (19, 16) 4 9 (2, 0) 6 (0, 0) 19 (17, 16) 0
This pattern is mostly maintained when calculating θ separately at synonymous and
nonsynonymous sites (Table 2.2), although the estimates for Π in UF3GT seem consid-
erably lower when considering the two classes of sites independently. Somewhat surpris-
ingly, synonymous and nonsynonymous Π are nearly equal at this locus, even though
it is expected that nonsynonymous sites would exhibit lower levels of polymorphism in
general (and this is the case for the measurements of θ
W
at UF3GT). In general, the
nonsynonymous levels of polymorphism seem either similar to or slightly lower than the
synonymous levels in the other fragments, with only one particularly noticeable excep-
tion: in DEFEN,nonsynonymousθ
W
inA. pubescens istwiceashighasthesynonymous
value, and is also much higher than the corresponding measure in A. formosa. This pat-
tern is only apparent in the θ
W
estimator, and not Π, which implies an excess of rare
variants at nonsynonymous sites in only A. pubescens. Because there are only 9 loci,
and each locus has a limited number of each type of site, it is difficult to draw many
conclusions from these types of comparisons.
9
Acetyl AP3 DEFEN Gapc H3 Heat LFY Pist UF3GT
θ θ
W
0.000 0.002 0.004 0.006 0.008 0.010 0.012 0.014
Acetyl AP3 DEFEN Gapc H3 Heat LFY Pist UF3GT
Π Π
0.000 0.002 0.004 0.006 0.008 0.010 0.012 0.014
Figure 2.1: Variation among levels of polymorphism for each species. The
sequences of each of the 9 fragments were grouped according to species, and θ
W
and Π
were estimated separately for each group. Grey bars represent estimates for A. formosa,
and black bars represent A. pubescens.
10
One pattern that stands out among all types of sites is that the estimates of θ are
strikingly correlated across species; as shall be demonstrated in the next paragraph, this
is because almost all variation is shared.
Table 2.2: Levels of polymorphism for synonymous and nonsynonymous sites.
Note that LFY is not listed in this table because it did not contain any nonsynonymous
sites.
Synonymous Nonsynonymous
θ
W
Π θ
W
Π
Fragment A.f A.p A.f A.p A.f A.p A.f A.p
Acetyl 0.0010 0.0011 0.0013 0.0013 0.0013 0.0022 0.0012 0.0009
AP3 0.0037 0.0041 0.0022 0.0017 0 0 0 0
DEFEN 0.0046 0.0034 0.0026 0.0021 0 0.0079 0 0.0015
Gapc 0.0061 0.0049 0.0018 0.0014 0.0118 0.010 0.0016 0.0042
H3 0.0069 0.0087 0.0015 0.0021 0.0092 0.0081 0.0029 0.0044
Heat 0.0063 0.0073 0.0038 0.0043 0.0017 0.0035 0.0006 0.0012
Pist 0.0023 0.0009 0.0003 0.0001 0.0010 0 0.0002 0
UF3GT 0.0215 0.0211 0.0053 0.0047 0.0109 0.0101 0.0055 0.0043
Comparing the minor allele counts for each species revealed that few high frequency
SNPs correspond to species-specific polymorphisms (Figure 2.2 and Table 2.1). In fact,
at sites with a minor allele frequency >5%, there were more than twice as many shared
polymorphisms (61) as species-specific polymorphisms (24). There were no fixed dif-
ferences in any of the 9 sequences. Wright’s F
ST
, a measure of genetic differentiation
between groups that ranges from 0 to 1 (where 0 means no differentiation among groups
and1indicatescompletedifferentiation), wasestimatedonaveragetobe0.0388between
the two species. A 95% confidence interval for this value is between 0.0383 and 0.0393,
which is quite low, but statistically different from zero. Despite the fact that the confi-
dence interval does not overlap with zero, when populations were randomly assigned to
2 groups regardless of species, the results were very similar: a mean F
ST
of 0.0349 with
a 95% C.I. between 0.0344 and 0.0353. Although this estimate is technically statistically
different from the original estimate, it also shows that even randomized groupings can
11
achieveF
ST
>0, and taken together these results suggest that very little of the observed
differentiation is due to species differences.
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● 0 5 10 15 20
0 5 10 15
Minor Allele Count in A. formosa
Minor Allele Count in A. pubescens
Figure2.2: Comparisonofminorallelecountsin A. formosa and A.pubescens.
The horizontal line represents the mean allele count in A. pubescens, while the vertical
line represents the mean allele count in A. formosa. Point size reflects the number of
comparisons at that point. Species-specific polymorphisms correspond to the points
along either the very bottom or the far left of the plot. All other sites correspond
to a shared polymorphism. There are no fixed differences. The average minor allele
frequency for any species-specific polymorphism was 0.105, while the average frequency
for any shared allele was 0.424.
2.2 Population Structure
Under the na¨ ıve assumption of 2 clusters, STRUCTURE (described in section 2.5) was
unable to cluster individuals according to species (Figure 2.3). In fact, the most likely
number of clusters appeared to be around 11 (Figure 2.4), based on when the estimated
probability and the average clusteredness stopped (consistently) increasing (Figures 2.5
and 2.6). Although the pattern of clustering does not correspond well to the geography
of the sample populations, it does not seem to be entirely random, especially among
the more well-defined clusters (where individuals tend to have membership coefficients
>0.5). Pairsofindividualsfromthesamepopulationtendedtoclustertogether≈15%of
the time, whereas pairs of individuals from different populations only clustered together
12
Table 2.3: Estimates of recombination
rate for each fragment.
Fragment Name R
M
ρ
Acetyl 1 0.081
AP3 11 0.003
Defen 8 0.133
Gapc 12 0.004
H3 3 0.00
Heat 6 0.126
LFY 4 0.006
Pist 4 0.001
UF3GT 15 0.023
Combined 0.009
≈9.5% of the time (p=0.00035 in χ
2
test). Similarly,≈11% of same-species pairs were
found in the same cluster, whereas only ≈8% of different-species pairs were clustered
together (p=0.03).
2.3 Isolation by Distance in the Aquilegia genus
In order to determine whether a high degree of shared polymorphism and a low level of
populationstructurewascommonintheAquilegia genusoruniquetoA. formosa andA.
pubescens, additionalsequencingwasperformedon2ofthe9originalgeneregions(Gapc
and UG3GT) in a broader sample of 32 Aquilegia species (including A. formosa and A.
pubescens) [80]. Although the majority of these individuals are also native to western
North America, the new sample encompasses an assortment of habitat types, pollinator
syndromes, and floral morphologies that is more representative of the Aquilegia genus
as a whole. The mean estimate of F
ST
between all pairs of populations was 0.247. In
addition, F
ST
was re-estimated between A. formosa and A. pubescens based on only
the two fragments sequenced for every other species and found to be 0.095. Although
this value is slightly higher than the original estimate, it is still much lower than the
estimate for most pairs of Aquilegia species. To further investigate the relationships
13
A. pubescens Piute Pass Trail, CA
A. pubescens Piute Pass Trail, CA
A. pubescens Piute Pass Trail, CA
A. formosa Upper Lehman Creek, NV
A. pubescens Morgan Pass, CA
A. pubescens Lamarck Trail, CA
A. pubescens Piute Pass Trail, CA
A. pubescens Lamarck Trail, CA
A. pubescens Piute Pass Trail, CA
A. formosa Bishop Creek, CA
A. formosa Upper Lehman Creek, NV
A. formosa Upper Lehman Creek, NV
A. formosa Upper Lehman Creek, NV
A. pubescens Lamarck Trail, CA
A. pubescens Lamarck Trail, CA
A. pubescens Piute Pass Trail, CA
A. pubescens Lamarck Trail, CA
A. pubescens Lamarck Trail, CA
A. pubescens Lamarck Trail, CA
A. pubescens Lamarck Trail, CA
A. formosa Las Vegas, NV
A. formosa White Mountains, CA
A. pubescens Piute Pass Trail, CA
A. formosa Las Vegas, NV
A. formosa Cascades, WA
A. formosa Las Vegas, NV
A. formosa Bishop Creek, CA
A. pubescens Piute Pass Trail, CA
A. formosa Upper Lehman Creek, NV
A. formosa Upper Lehman Creek, NV
A. formosa Fresno, CA
A. formosa Fresno, CA
A. formosa Ventura, CA
A. formosa Ventura, CA
A. formosa Ventura, CA
A. formosa Ventura, CA
A. formosa Bishop Creek, CA
A. formosa Fresno, CA
A. formosa Upper Lehman Creek, NV
A. pubescens Morgan Pass, CA
A. pubescens Piute Pass Trail, CA
A. formosa Cascades, WA
A. formosa Cascades, WA
A. formosa Cascades, WA
A. formosa Cascades, WA
A. formosa Fresno, CA
A. formosa Fresno, CA
A. pubescens Lamarck Trail, CA
A. pubescens Lamarck Trail, CA
A. pubescens Lamarck Trail, CA
A. pubescens Piute Pass Trail
A. formosa Upper Lehman Creek, NV
A. pubescens Piute Pass Trail, CA
A. pubescens Lamarck Trail, CA
A. pubescens Piute Pass Trail, CA
A. pubescens Lamarck Trail, CA
A. pubescens Piute Pass Trail, CA
A. pubescens Lamarck Trail, CA
A. pubescens Lamarck Trail, CA
A. formosa Po Island, British Columbia
A. formosa Po Island, British Columbia
A. formosa Po Island, British Columbia
A. formosa Upper Lehman Creek, NV
A. formosa Po Island, British Columbia
A. formosa Bishop Creek, CA
A. formosa Po Island, British Columbia
A. formosa Upper Lehman Creek, NV
A. pubescens Morgan Pass, CA
A. pubescens Morgan Pass, CA
A. formosa White Mountains, CA
A. formosa White Mountains, CA
A. formosa White Mountains, CA
A. formosa White Mountains, CA
A. pubescens Lamarck Trail, CA
A. formosa Alaska
A. formosa Bishop Creek, CA
A. formosa Las Vegas, NV
A. formosa Las Vegas, NV
A. pubescens Piute Pass Trail, CA
A. pubescens Piute Pass Trail, CA
Figure2.3: STRUCTUREcannotclusterindividualsaccordingtospecies. Each
individual is indicated by a thin line, where the two colors represent the estimated
membership coefficients for the 2 clusters. The clusteredness score for this plot was
estimated as 0.26.
14
A. formosa Upper Lehman Creek, NV
A. formosa Fresno, CA
A. pubescens Piute Pass Trail, CA
A. pubescens Piute Pass Trail
A. pubescens Piute Pass Trail, CA
A. pubescens Lamarck Trail, CA
A. formosa White Mountains, CA
A. pubescens Lamarck Trail, CA
A. formosa White Mountains, CA
A. pubescens Lamarck Trail, CA
A. pubescens Piute Pass Trail, CA
A. pubescens Lamarck Trail, CA
A. pubescens Lamarck Trail, CA
A. pubescens Lamarck Trail, CA
A. formosa Ventura, CA
A. pubescens Piute Pass Trail, CA
A. pubescens Morgan Pass, CA
A. formosa Cascades, WA
A. formosa Bishop Creek, CA
A. pubescens Piute Pass Trail, CA
A. pubescens Piute Pass Trail, CA
A. pubescens Lamarck Trail, CA
A. pubescens Piute Pass Trail, CA
A. formosa Upper Lehman Creek, NV
A. formosa Upper Lehman Creek, NV
A. pubescens Piute Pass Trail, CA
A. formosa Bishop Creek, CA
A. pubescens Lamarck Trail, CA
A. formosa Upper Lehman Creek, NV
A. formosa Upper Lehman Creek, NV
A. pubescens Lamarck Trail, CA
A. pubescens Lamarck Trail, CA
A. pubescens Lamarck Trail, CA
A. pubescens Lamarck Trail, CA
A. formosa Po Island, British Columbia
A. formosa Cascades, WA
A. formosa Upper Lehman Creek, NV
A. formosa Po Island, British Columbia
A. formosa Po Island, British Columbia
A. formosa Po Island, British Columbia
A. formosa Fresno, CA
A. formosa Upper Lehman Creek, NV
A. formosa Alaska
A. pubescens Piute Pass Trail, CA
A. formosa White Mountains, CA
A. formosa White Mountains, CA
A. formosa Las Vegas, NV
A. formosa Las Vegas, NV
A. formosa Upper Lehman Creek, NV
A. formosa Fresno, CA
A. formosa Bishop Creek, CA
A. formosa Ventura, CA
A. formosa Las Vegas, NV
A. formosa Bishop Creek, CA
A. pubescens Morgan Pass, CA
A. formosa White Mountains, CA
A. formosa Fresno, CA
A. formosa Ventura, CA
A. pubescens Morgan Pass, CA
A. pubescens Piute Pass Trail, CA
A. formosa Upper Lehman Creek, NV
A. pubescens Piute Pass Trail, CA
A. pubescens Piute Pass Trail, CA
A. pubescens Piute Pass Trail, CA
A. formosa Fresno, CA
A. formosa Po Island, British Columbia
A. formosa Bishop Creek, CA
A. pubescens Lamarck Trail, CA
A. formosa Cascades, WA
A. pubescens Lamarck Trail, CA
A. pubescens Piute Pass Trail, CA
A. formosa Upper Lehman Creek, NV
A. formosa Ventura, CA
A. pubescens Lamarck Trail, CA
A. formosa Las Vegas, NV
A. pubescens Morgan Pass, CA
A. formosa Las Vegas, NV
A. pubescens Lamarck Trail, CA
A. formosa Cascades, WA
A. formosa Cascades, WA
Figure 2.4: Inferred population structure for 80 Aquilegia individuals. The
results from STRUCTURE are plotted for K=11, which had an average clusteredness
score of ≈ 0.52. Each individual is represented by a thin horizontal line, with corre-
sponding population and species information given on either side.
15
● ● ● ● ● ● ● ● ● ● ● ● ● ● 2 4 6 8 10 12 14
−6500 −6000 −5500 −5000 −4500
Number of Clusters (K)
Estimated Ln Probability
Figure 2.5: Probability of Different K Estimates: The estimated log probability of
the data (as calculated by STRUCTURE) is plotted against different K values. For each
K value, STRUCTURE was run 3 times, and the plotted value is the average of those 3
runs.
● ● ● ● ● ● ● ● ● ● ● ● ● ● 2 4 6 8 10 12 14
0.30 0.35 0.40 0.45 0.50 0.55
Number of Clusters (K)
Average Clusteredness
Figure 2.6: Average Clusteredness for Different K Values. For each K value, the
average clusteredness measures the extent to which each individual belongs to a single
cluster rather than to multiple clusters, so the higher the clusteredness the “better” the
clusters.
16
among these species, a simple genetic distance tree was generated based on combined
informationfrombothloci(Figure2.7). Inthistree,therelationshipbetweenA. formosa
and A. pubescens does not reflect the true species tree (Figure 2.8), whereas members
of other Aquilegia species seem to cluster together more frequently. European and Asia
individuals form the most basal part of the phylogeny, suggesting that there may be a
relationship between geographic distance and genetic differentiation among these taxa.
This hypothesis was formally tested using a Mantel test to determine if there were a
correlation between F
ST
and physical separation across all pairs of populations with at
least 5 individuals (Figure 2.9). Results of this test indicated that geographic distance
didindeedhaveasignificantrelationshipwithgeneticdifferentiationwithintheAquilegia
genus(r =0.619,P =0.0008)–moresothananyotherfactorthatwasexamined(includ-
ing pollinator syndrome and habitat type)(Figure 2.10). However, on a more local scale,
there is no evidence for isolation by distance either within or between A. formosa or A.
pubescens populations (r = 0.171, P = 0.307) (Figure 2.11). This finding is not only in
agreement with the results discussed in Section 2.2 and Section 2.1, but also suggests
that these two species are even more closely related than other members of this genus.
2.4 Linkage Disequilibrium and Recombination Rate
Linkage disequilibrium (LD) was not extensive in any of the 9 regions that were
sequenced, with average correlation coefficients (r
2
) ranging between 0.1 and 0.2 for
most of the fragments. When values of r
2
are plotted against physical distance between
SNPs, the relationship is weak (Figure 2.12). The fragments with the highest levels of
polymorphism show evidence for a rapid decay of LD (within about 1kb or less). The
combined fragment data show low LD values overall, and the estimate of ρ was 0.009,
which is higher than estimates of ρ in humans [55], suggesting a relatively high rate of
recombination in these species (Table 2.3).
17
Alaska
British Columbia
CA!Morgan Pass
NV!Upper Lehman Creek
CA!Fresno
CA!Fresno
CA!Lamarck Trail
CA!Lamarck Trail
CA!Lamarck Trail
NV!Las Vegas
CA!Piute
CA!Piute
British Columbia
CA!Bass Lake
NV!Upper Lehman Creek
NV!Upper Lehman Creek
British Columbia
NV!Stella Lakes
WY!Jackson Hole
British Columbia
British Columbia
CA!Fresno
NV!Las Vegas
CA!Lamarck Trail
CA!Piute
CO!Loveland Pass
CO!Loveland Pass
WY!Friend Camp
CA!Morgan Pass
CA!Lamarck Trail
NV!Upper Lehman Creek
CA!Lamarck Trail
British Columbia
CA!Bishop Creek
NV!Las Vegas
CA!Piute
CA!Lamarck Trail
CA!Lamarck Trail
NV!Upper Lehman Creek
CA!White Mountains
CA!White Mountains
CA!Morgan Pass
CA!Lamarck Trail
CA!Piute
CA!Morgan Pass
WA!Cascades
CA!Piute
CA!Lamarck Trail
WA!Cascades
CA!Morgan Pass
CA!Lamarck Trail
CA!Piute
CA!Ventura
CA!Lamarck Trail
NV!Las Vegas
NV!Upper Lehman Creek
CA!Piute
CA!White Mountains
CA!Piute
CO!Rifle Falls
CO!Rifle Falls
CO!Rifle Falls
CO!Escalante
CO!Escalante
CO!Escalante
NV!Upper Lehman Creek
AZ!Barboquivari
AZ!Barboquivari
TX!Capote Falls
NM!Sitting Bull Falls
NM!Sitting Bull Falls
NM!Sitting Bull Falls
NM!Sitting Bull Falls
UT!High Creek
UT!High Creek
TX!Maple
TX!Maple
TX!Maple
TX!Maple
CA!Piute
NV!Stella Lakes
NV!Charleston Peak
NV!Charleston Peak
NV!Charleston Peak
UT!Zion
UT!Zion
UT!Zion
NV!Stella Lakes
TX!Maple
El Salto
El Salto
CO!Loveland Pass
WY!Sugarloaf
UT!Tony Grove
Mexico
Mexico
Mexico
TX!Capote Falls
NM!Hyde Memorial
NM!Hyde Memorial
CA!Piute
UT!Zion
WY!Sugarloaf
CA!Porter Camp
AZ!Sawmill
UT!High Creek
CA!Piute
WY!Sugarloaf
WY!Sugarloaf
AZ!Barboquivari
NV!Charleston Library
NV!Charleston Library
NV!Stella Lakes
NV!Stella Lakes
UT!Tony Grove
NM!Hyde Memorial
CA!Bass Lake
NV!Upper Lehman Creek
CO!Escalante
NV!Charleston Library
NV!Charleston Library
British Columbia
British Columbia
WA!Cascades
CA!Piute
CA!Ventura
CA!Bishop Creek
CA!Lamarck Trail
NV!Upper Lehman Creek
British Columbia
CA!Lamarck Trail
CA!Fresno
CA!Bishop Creek
CA!Lamarck Trail
CA!Ventura
CA!Lamarck Trail
CA!Lamarck Trail
NV!Upper Lehman Creek
NV!Upper Lehman Creek
WY!Jackson Hole
NV!Las Vegas
NV!Upper Lehman Creek
CA!Ventura
CA!Lamarck Trail
UT!Zion
CA!Bishop Creek
CA!Bishop Creek
CA!Piute
CA!Piute
CA!Piute
WY!Jackson Hole
WY!Friend Camp
CO!Escalante
WA!Cascades
NV!Upper Lehman Creek
Alberta
Ontario
Ontario
Vancouver
Vancouver
WY!Medicine Wheel
CO!Pikes Peak
CO!Pikes Peak
CO!Pikes Peak
CO!Pikes Peak
CO!Pikes Peak
Vancouver
Vancouver
Ontario
Ontario
Vancouver
WY!Medicine Wheel
Mexico
Mexico
WY!Jackson Hole
CA!Piute
CA!Piute
CA!Piute
CA!Lamarck Trail
AZ!Sawmill
AZ!Horse Spring
NM!Hyde Memorial
TX!Capote Falls
TX!Capote Falls
AZ!Barboquivari
TX!Capote Falls
Eurasia
Pyrenese
CA!Piute
WY!Jackson Hole
UT!Tony Grove
UT!High Creek
NV!Charleston Peak
UT!Tony Grove
CA!Lamarck Trail
CA!Piute
British Columbia
NV!Charleston Peak
El Salto
NM!Ash Canyon
NM!Ash Canyon
Eurasia
Eurasia
Eurasia
Eurasia
Eurasia
Figure 2.7: Phylogeny for Two genomic sequences.
18
Figure2.8: TruePhylogenyforSeveralMembersofthe Aquilegia genus. Figure
adapted from the website of S. Hodges
.
2.5 Materials and Methods
2.5.1 Sample collection and preparation
Leaf tissue was collected from individual plants found in different locations along the
west coast of North America. Samples were taken from 40 individuals of each species,
for a total sample size of 80 individuals. A. formosa samples were taken from 9 different
populations, ranging from California, Nevada, Washington state, British Columbia, and
Alaska. The number of individuals in each of these populations varied between 1 and 10,
but most populations had 5 individuals. There were only 3 populations of A. pubescens,
andallofthemwerefromCalifornia. Therewerebetween4and16individualsineachof
these populations (see also Table 2.4 for a description of the sampling). Because the A.
pubescens populations were less geographically dispersed than the A. formosa samples,
there was some concern that A. pubescens might falsely appear to be less polymorphic
than A. formosa. However, the same level of polymorphism was found in both species,
so sampling bias was not an issue.
19
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● 0 1000 2000 3000
0.1 0.2 0.3 0.4 0.5 0.6
Distance (km)
Fst
Figure 2.9: Relationship between geographic distance and genetic distance.
Each dot represents a comparison between 2 populations of at least 5 individuals. For
populations where there were more than 5 individuals, estimates of F
ST
were boot-
strapped to ensure that the larger sample size did not cause any bias in the estimate.
DNAextractionswereperformedusingQiagen’sDNeasyPlantMiniExtractionKits.
Due to limited sample amounts, extracted DNA was used directly in only 5 out of the
9 amplifications (Acetyl, Defen, H3, LFY, and UF3GT). For the remaining 4 amplifi-
cations, the extracted DNA was first amplified using Qiagen’s REPLI-g Mini Kit and
corresponding whole genome amplification protocol.
Additionalleaveswerecollectedfromthirty-twoAquilegia taxa(includingA. formosa
and A. pubescens)[80]. Twenty-five of these are also native to North America, while the
remaining7arefoundinEuropeandAsia. Foreachspecies,between1and3populations
were sampled, with an average of 5 individuals per population (Table 2.4). The majority
of individuals came from western North America. DNA extractions were performed as
described above.
2.5.2 Fragment amplification and sequencing
NineshortregionsoftheAquilegia genomewereamplifiedintheoriginalsampleviaPCR
using 3’-UTR anchored primers (Table 2.5). These primers were originally designed by
20
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● 0 1000 2000 3000 4000
0.0 0.1 0.2 0.3 0.4 0.5 0.6
Distance (km)
Fst
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● (a) Within vs. Between Species
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● 0 1000 2000 3000 4000
0.0 0.1 0.2 0.3 0.4 0.5 0.6
Distance (km)
Fst
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● (b) Same vs. Different Pollinator Syndrome
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● 0 1000 2000 3000 4000
0.0 0.1 0.2 0.3 0.4 0.5 0.6
Distance (km)
Fst
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● (c) Same vs. Different Habitat Type
Figure 2.10: Other factors influencing F
ST
in Aquilegia. In all panels, red dots
indicate comparisons where both populations were the same for the factor being con-
sidered, while gray dots indicate comparisons where the two populations were different.
Panel (A) shows F
ST
vs distance both within and between species, with the green dia-
monds indicating comparisons between either A. formosa or A. pubescens and one of
the natural hybrid populations. Panel (B) shows F
ST
vs distance with the same and
different pollinator syndrome, while Panel (C) shows the same comparisons for habitat
type. 21
0 500 1000 1500
0.00 0.05 0.10 0.15 0.20 0.25
Distance (km)
Fst
● Figure2.11: Relationshipbetweengeographicdistanceandgeneticdistancefor
A. formosa and A.pubescens only. Black squares represent comparisons between
A. formosa and A. pubescens populations; gray triangles are comparisons among popu-
lations of A. formosa; white circles are comparisons among populations of A. pubescens.
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● 0 200 400 600 800 1000
0.0 0.2 0.4 0.6 0.8 1.0
Combined Fragment Data
Distance (base pairs)
R−squared
Figure 2.12: R
2
versus distance for the combined data. Only SNPs with a minor
allele frequency ≥ 10% were used in pairwise comparisons. Trend line was fitted as
described in the Methods section.
22
Table 2.4: Summary of Aquilegia samples used in this study.
Species Population Num. Ind. Sample
A. barnebyi Rifle Falls, CO 3 Second
A. brevistyla Alberta, Canada 5 Second
A. canadensis Kimbalton, Virginia 5 Second
A. canadensis Ontario, Canada 5 Second
A. chaplenei Sitting Bull Falls, NM 5 Second
A. chrysantha Ash Canyon, NM 5 Second
A. chrysantha El Salto, NM 4 Second
A. coerulea var. alpina Jackson Hole, WY 5 Second
A. coerulea var. coerulea Loveland Pass, CO 5 Second
A. coerulea var. coerulea Sugarloaf, WY 5 Second
A. coerulea var. coerulea Tony Grove, UT 5 Second
A. pinetorum Sawmill, AZ 5 Second
A. desertorum Horse Spring, AZ 1 Second
A. elegantula Hyde Memorial, NM 5 Second
A. eximia Porter Camp, CA 2 Second
A. flavescens High Creek, UT 5 Second
A. formosa Alaska 1 First
A. formosa Bass Lake, CA 5 Second
A. formosa Bishop Creek, CA 5 First
A. formosa Cascades, WA 5 First
A. formosa Fresno, CA 5 First
A. formosa Las Vegas, NV 5 First
A. formosa Po Island, British Columbia 5 Second
A. formosa Po Island, British Columbia 5 First
A. formosa Upper Lehman Creek, NV 5 Second
A. formosa Upper Lehman Creek, NV 10 First
A. formosa Ventura, CA 4 First
A. formosa White Mountains, CA 5 First
A. hinckleyana Capote Falls, TX 5 Second
A. jonesii Medicine Wheel, WY 2 Second
A. laramiensis Friend Camp, WY 2 Second
A. longissima Baboquivari, AZ 4 Second
A. longissima Maple. TX 5 Second
A. micrantha Escalante, CO 5 Second
A. pubescens Lamarck Trail, CA 16 First
A. pubescens Morgan Pass, CA 4 First
A. pubescens Piute Pass Trail, CA 5 Second
A. pubescens Piute Pass Trail, CA 15 First
A. saximontana Pikes Peak, CO 5 Second
23
Table 2.4, continued
Species Population Num. Ind. Sample
A. scopulorum Charleston Peak, NV 5 Second
A. shockleyi Charleston Library, NV 5 Second
A. skinneri Mesa del Campanero, Sonora, MX 5 Second
A. triternata Zion, UT 5 Second
A. caucausica Europe 1 Second
A. flabellata Japan 3 Second
A. fragrans Europe 1 Second
A. glandulosa Asia 1 Second
A. olympica Europe 2 Second
A. pyrenaica Europe 2 Second
A. viridiflora Siberia 3 Second
Hybrids Lamarck Lakes, CA 11 First
(formosa and pubescens)
Hybrids Piute, CA 5 First
(formosa and pubescens)
Sp. nov. Stella Lakes, NV 5 Second
Whittall et al. [81] to reconstruct a species-level phylogeny for several members of the
Aquilegia genus (including A. formosa, but excluding A. pubescens). None of these
regions are expected to be involved in the evolution of reproductive barriers. Two of the
9 regions were also amplified in the broader sample of 32 species (Gapc and UF3GT).
All of the sequences contained some non-exonic DNA (Table 2.6).
All PCR amplifications were done in a total volume of 25μL, with 20μL Promega
PCR Master Mix (2X: 50 units per mL of Taq polymerase, 400μM dATP, 400μM dGTP,
400μM dCTP, 400μM dTTP, 3mM MgCl
2
), 3μL of forward and reverse primers (10μM
each), and approximately 20 ng of DNA template. Although the annealing temperature
variedslightlyamongprimerpairs,thecyclingconditionsweregenerallyasfollows: 92
◦
C
for 2 minutes, followed by 35 cycles of: 92
◦
C for 45 seconds, 61
◦
C for 30 seconds, 72
◦
C
for 1.5 minutes, and a final extension step at 72
◦
C for 10 minutes.
Sequencingfortheoriginalsampleof80individualswasperformedinbothdirections
using the Beckman-Coulter CEQ 2000 platform. Purifications and sequencing reactions
were all done as recommended by the Beckman-Coulter protocols. PCR products were
24
Table 2.5: Primer pairs used to amplify the 9 nuclear loci.
Locus Frag. Size (bp) Primers (F=top; R=bottom)
Acetyl-CoA 577 ATTCGCGGAGCTACATGATA
carboxylase (Acetyl) CCTACTGCTACTTTCAACAATCAAC
Defensin protein (DEFEN) 683 GCAACATGCGTCTAGTTTCAG
GAACCACGAAGGTGACCCT
Glyceraldehyde-3-phosphate 1097 GTCTGAGGGCAAACTGAAGG
dehydrogenase (GAPC) AAACCTGAAGCAGCAATAGGA
Histone H3 358 CAAACTTCCCTTCCAACGTC
AACTTCCGATATATTTCATTCATTG
HSP70-1 (HEAT) 1281 TCTTCAGGGAGAGAGAGAGTTTG
ATTACTTCCCCACCATCAGG
Apetala-III (AP3) 1093 TGAGTCTGTGAAACTTGTTCGGG
GCAATGCGAATAGCAATGCC
LFY 575 CCCAACCAAGGTATGAATA
CCTGAATTGCATGTCGATACAC
Pistalla-AqaPI-1 (Pist) 1200 CGAACTCAGGCACTTGAAGG
GCATTGTTGAATGTTGATACACTCT
Glycosyl transferase (UF3GT) 473 GAGGAAGCTTTGCCAGAGG
AAATGCGACACTGCGACATA
purified using Promega’s Wizard MagneSil PCR Clean-Up System. Eight microliters
of purified template were mixed with 1μL CEQ 10X Buffer, 1μL CEQ QuickStart Mix,
2.8μL water, and 0.25μL of either forward or reverse primer (for a total reaction volume
of 13μL). The sequencing reaction mixtures were then subjected to the following cycling
conditions: 96
◦
C for 20 seconds, 50
◦
C for 20 seconds, and 60
◦
C for 4 minutes for a total
of 40 cycles, followed by holding at 4
◦
C. The reaction products were cleaned up using
the Beckman-Coulter protocol for “Ethanol Plate Precipitation in a CEQ sample plate,”
and then finally loaded into the CEQ 2000 for sequencing. Sequencing for the broader
sample was performed in the Hodges lab at UC Santa Barbara on the Li-Cor System.
25
Table 2.6: Positions of introns, exons, and UTRs in Each Locus
Locus Exon Positions Intron Positions UTR Positions
Acetyl 1...389, 485...542 390...484 543...577
DEFEN 1...38, 389...560 39...388 561...683
GAPC 1...59, 231...318, 844...888 60...230, 319...843 889...1097
H3 1...200 ... 201...358
HEAT 1...192, 477...482, 930...773 193...476, 483...929 774...1281
AP3 1...6, 120...163, 542...587, 7...119, 164...541, 833...1093
678...832 588...677
LFY ... 1...575 ...
Pist 1...762 763...945 946...1200
UF3GT 1...424 ... 425...473
2.5.3 Sequence alignment and editing
Sequences obtained from the CEQ 2000 were aligned using phredPhrap [13, 14], and
visualized in Consed [20]. All alignments were edited manually with the aid of MABCW
(program written by T. Hu; scripts and more information available upon request). The
indel polymorphisms that were identified were all relatively short, and only two alleles
were observed at each of these sites. It was not possible to characterize individuals that
were heterozygous at these sites, so they were treated as missing data. For homozygous
individuals, indels were analyzed as biallelic SNPs.
For each fragment, the set of segregating sites was identified using alignments of all
sequences from both species. The sites in this set were then subsequently characterized
as either exclusive to one species or shared based on whether or not they were still
segregating in an alignment of sequences from only one species. At each SNP position,
the derived allele was determined by using a draft assembly of the Aquilegia coerulea
(Goldsmith)genomeasanoutgroup(JointGenomeInstitute(JGI)Aquilegia Sequencing
Project, unpublished data).
For the purpose of linkage disequilibrium analyses, haplotypes were reconstructed
using PHASE 2.0.2 [65, 64]. For all other analyses, (estimation of θ, F
ST
, MIMAR, and
population structure), the un-phased genotype data was used directly.
26
2.5.4 Analysis
Using in-house scripts (available upon request), Watterson’s estimator (θ
W
) and the
average number of pairwise differences (Π) were determined for each of the 9 sets of
sequences and then scaled by the length of the sequence in order to get a per base pair
value. ThereadingframeforeachfragmentwasassumedbasedonalignmentwithcDNA
sequences available in Genbank (accession numbers: DQ286961, DQ224264, DQ224271,
DQ217409, DQ286960, DQ224258, AY162852, and DQ286959). Estimates of θ for dif-
ferent classes of sites were scaled by the total number of silent sites or nonsynonymous
sites in each sequence. The number of silent sites (S) and the number of nonsynonmous
sites (N) were calculated based on a simple Jukes-Cantor model of substitution [30],
with the following equations: S =
L
2
3
+L
4
, N =
2L
2
3
+L
0
, where L
0
is the number of
non-degenerate sites,L
2
is the number of twofold degenerate sites, andL
4
is the number
of fourfold degenerate sites.
Estimates of Wright’s F
ST
were calculated based on estimates of Π [29] using the
following equation:
F
ST
=
Π
between
−Π
within
Π
between
whereΠ
between
referstotheaveragepairwisedifferencebetweenindividualsfromdifferent
species,andΠ
within
istheaveragepairwisedifferencewithinspecies. Confidenceintervals
were obtained by using 10,000 bootstrap replicates.
Population structure was inferred directly from the sequence data using the program
STRUCTURE 2.0, which implements a model-based clustering approach [54]. STRUC-
TUREwasrununderthe“linkagemodel”with“correlatedallelefrequencies.”Specifying
correlated allele frequencies enhances the ability of the algorithm to detect distinct clus-
ters even among a sample of very closely related populations [57], which is well suited
to the Aquilegia data set. Although geographic sampling information was available, ini-
tial STRUCTURE runs suggested that geographic location did not correspond well with
the genetic data, so the “prior population information” model was not used to assist in
27
clustering. The program was run with a burn-in length of 50,000 and a run length of
20,000. This was done several times for each K value (ranging from 2 to 15) in order to
ensure that results were consistent. Plots of the STRUCTURE output were generated
using distruct [56]. The average “clusteredness” of individuals was calculated for each
STRUCTURE run according to the equation presented by Rosenberg et al. [57].
Analysis of Isolation By Distance (IBD) was performed using a Mantel test [44]
with 10,000 replications as implemented by the R package “ade4” [70, 12]. The genetic
distancematrixwascomposedofestimatesforF
ST
whilethegeographicdistancematrix
was measured in kilometers between populations.
For genetic distance calculations, each SNP was treated as a separate marker, and
the distance between individuals was calculated using the program “dnadist” from the
PHYLIP package [15]. Trees were generated using the R package “cluster” [42].
Linkage disequilibrium (LD) between SNPs was quantified using r
2
, the squared
correlation coefficient. For each fragment, r
2
was plotted as a function of the distance
between SNPs (measured in base pairs). The population recombination parameter (ρ)
was estimated by fitting the equation given in [79, 9]:
E(r
2
)=
10+Γ
22+13Γ+Γ
2
×

1+
(3+Γ)(12+12Γ+Γ
2
)
n(22+13Γ+Γ
2
)
2

whereΓ=ρ∗(distance). Forallanalysesofrecombination, lowfrequency(MAF≤10%)
polymorphisms were removed, since they provide little information about the overall
pattern of LD.
2.6 Summary
The genetic data generated by direct sequencing of 9 nuclear loci in 80 individuals from
multiplepopulationscouldnotdistinguish A. formosa from A. pubescens. Notonlywere
valuesofθ strikinglysimilaracrossspeciesforeveryfragment, butestimatesofF
ST
were
also extremely low, indicating that almost all polymorphism is shared between species.
28
This is a remarkable finding given that these two species are strongly differentiated both
ecologically and morphologically.
The patterns of population structure were also unclear; geographic distance between
populations had a clear correlation with genetic differentiation in the broad sample
of North American Aquilegia taxa, but there is not a clear relationship when only A.
formosa and A. pubescens are examined on a more local scale. At the same time,
the clustering of individuals in STRUCTURE does not seem entirely random, with two
individualsbeingmorelikelytoclustertogetheriftheyarefromthesamespeciesandthe
same population than if they are not. It is possible that these results are a reflection of
a pre-exisiting population structure in the common ancestor, or that migration between
populationshasmadethestructurehardertodiscern. Thisraisesthequestionofwhether
or not gene flow has played a role in shaping these species, an issue which will be
addressed in Chapter 3.
Theresultsofthisstudyindicatethatahigh-throughputwholegenomere-sequencing
approachmightbeanidealwaytofindthelociresponsibleforreproductiveisolationand
gain a clearer understanding of how speciation has occurred in Aquilegia. A relatively
recent scan of genome-wide patterns of interspecific differentiation in two species of
European oaks led to the identification of a few genomic regions which seem to underlie
species divergence [59]. Like Aquilegia, these oak species were closely related and highly
interfertile, despite exhibiting significant differences in ecology and morphology. The
overall low levels of interspecific variation in these species facilitated the identification
of highly differentiated regions, and based on the finding that genetic differentiation is
incredibly low at neutral loci in Aquilegia, it is reasonable to expect that a similar scan
of the Aquilegia genome would be equally effective.
29
Chapter 3
Sympatric vs. Allopatric
Speciation
The relative importance of natural selection versus geographic isolation in the spe-
ciation process has been debated since Darwin first published the Origin of Species
[11]. Although Darwin seemed to favor the role of natural selection over biogeography,
the prevailing view for the last century has been that reproductive isolation is almost
always caused by geographic isolation followed by random genetic drift (allopatric spe-
ciation), and that speciation caused by selection in the face of gene flow (sympatric
speciation) could only happen in extreme circumstances, if at all [7, 76]. Recently, how-
ever, there has been resurgent interest in sympatric speciation, and many studies in
the last ten years have focused on the role of gene flow during the speciation process
[24, 67, 8, 85, 3, 43, 50, 60]. These studies have shown that adaptive differences between
species can be maintained even in the face of significant amounts of introgression, espe-
cially if only a few genes or genomic regions control the traits that lead to reproductive
isolation [67].
Genome-wide analyses of many species have shown that levels of introgression can
vary across the genome, with divergent selection playing an active role in preventing
gene flow at the loci underlying adaptive traits, but not acting at other areas in the
genome [67, 24, 75]. Sympatric speciation can be difficult to prove, however, as incipient
species will also show varying levels of differentiation across the genome, with the most
differentiated regions also being the most likely to contain genes that restrict random
30
mating [24, 3]. These species can appear almost identical at many loci, even in the
complete absence of genetic exchange.
In Chapter 2, it was shown that the genetic variation found at nuclear loci failed
to distinguish A. formosa and A. pubescens, even though these two species are both
morphologically and ecologically distinct. Several of the sympatric speciation studies
mentioned earlier uncovered the same phenomenon. Different species of wild sunflow-
ers exhibit strong ecological differentiation, but it has been found that there are few
fixed differences between the species, despite very high levels of intraspecific variation
(higher than what we observed in Aquilegia) [67]. Hybridization also occurs between
these species, and there is evidence for long-term introgression since their divergence one
million years ago [67]. Gene flow has also played a role in shaping the patterns of genetic
divergence among species in the Hawaiian silversword alliance, which (like Aquilegia) is
another example of an adaptive radiation in plants [38]. Finally, African cichlid fishes
represent one of the most dramatic examples of an adaptive radiation, and many of the
morethan2,000uniquespeciesinthisgrouphavearisenviasympatricspeciationandare
still capable of forming viable hybrid offspring, despite many ecological, morphological,
and behavioral differences [34].
Asintheaboveexamples,itisknownthathybridzonesformbetweenA. formosa and
A. pubescens [26, 4, 21]. There are also some genetic markers that suggest introgression
beyond the hybrid zones [26], which makes it tempting to speculate that gene flow
between the species has been occurring since their divergence. In this part of the study,
animplementationoftheisolation-migrationmodel[23,2]wasusedtoproduceestimates
ofthedivergencetimeandgeneflowbetweenspecies. Theresultswerepublishedaspart
of a paper in PLoS One [6], and are detailed in this chapter.
31
3.1 The Isolation-Migration Model
Whentwospeciesexhibithighlevelsofsharedpolymorphism,asinthecaseofA.formosa
and A. pubescens, this can either be the result of persistent ancestral polymorphisms
or continued genetic exchange between them. If divergence was very recent, then it
is not unreasonable to expect that most of the genetic variation in each species was
inherited from a common ancestor, and so they might look very similar even without
gene flow. On the other hand, if interbreeding is prevalent, then the genomes of two
species could still look genetically identical even if divergence happened a long time ago.
These two scenarios respectively reflect allopatric and sympatric speciation, and should
be regarded as the two extreme ends of a spectrum, where there are varying levels of
genetic exchange in the middle [3, 18]. In many cases, including this one, two species are
partially geographically isolated (in parapatry), and it seems more intuitive to imagine
that what is going on is something between allopatric and sympatric speciation [3].
The purpose of the Isolation-Migration (IM) model is to simultaneously estimate
the divergence time and the level of gene flow for two populations using a probabilis-
tic approach [23, 24]. Using multi-locus DNA sequence data, this method estimates 6
population parameters: the ancestral mutation rate (θ
A
), the mutation rates for the two
current populations (θ
1
andθ
2
), the time since split (t), the migration rate from popula-
tion 1 into population 2 (m
1
), and the migration rate from population 2 into population
1(m
2
)(Figure3.1). Startingfromuser-definedrangesofeachparameter, themodeluses
Markov Chain Monte Carlo (MCMC) simulations to estimate a set of genealogies from
the data, and then concurrently determines the likelihood of these genealogies based on
the polymorphism data. This framework accommodates the parapatric conditions found
in many systems (including Aquilegia), and can offer a more realistic model of speciation
without having to definitively prove either allopatry or sympatry (which is difficult to
do).
32
Figure 3.1: The isolation with migration (IM) model with 6 parameters.
Adapted from [23].
MIMAR [2] is a modified version of the original IM model which can incorporate
information about intralocus recombination rates, and this is the program that was
used on the Aquilegia data. For these analyses, the recombination rate was set at r =
0.009, basedontheestimatedpopulationrecombinationratefromlinkagedisequilibrium
data (Section 2.4). θ
1
, θ
2
, and θ
A
were all sampled from a uniform prior distribution
U[0.002,0.009], a range which spans the mean estimates of θ found in Section 2.1. The
time since split, T, measured in generations, was sampled from the prior distribution
U[0,100000]. The program was run for 1.1×10
5
recorded steps, and 1×10
4
burn–in
steps, which was found to achieve adequate mixing of the Markov chain.
3.2 A Simple Model with No Migration
Since gene flow allows species to remain genetically similar over time, running the IM
model under a scenario with no gene flow should produce an estimate of the earliest
possible divergence time. When MIMAR was run with the migration rate fixed at 0, the
time since the split between A. formosa and A. pubescens is estimated as approximately
0.062 in coalescent time units (Figure 3.2). If the mutation rate (μ) is assumed to be
6×10
−9
(based on an estimation of the average substitution rate in nuclear DNA in
33
plants [82]), then the equivalent number of generations can be calculated through the
following equations:
T
gen
=T
coal
∗4N
1
=T
coal
∗
θ
μ
which results in an estimate of 55,784.5 generations. If the actual mutation rate in
Aquilegia is higher than was assumed, then the estimated number of generations since
the split will be lower, and if the actual mutation rate is lower, than the number of
generations will be higher. The generation time in Aquilegia is not known, but a very
rough estimate can be calculated as 10 years, based on the observation that the plants
seem to produce seeds in the wild for about 20 years. If we assume the generation time
is around 10 years, then the MIMAR results suggest that A. formosa and A. pubescens
diverged 557,845 years ago.
3.3 Migration Rate Estimates
Before adding the migration parameters into the IM model, a simpler way to estimate
migration is to use Wright’s equation [83] for an n-island population model, which is
based on F
ST
:
F
ST
=
1
1+4N
e
m
whereM =4N
e
m,orthenumberofmigrantsbetweenpopulationspergeneration. Using
the estimates of F
ST
obtained in Section 2.1, M = 6. Because it is generally accepted
that A. formosa and A. pubescens have diverged recently (and MIMAR supports this
theory), it is reasonable to assume that at least some of the shared variation is due
to ancestral polymorphism, and is not solely the result of gene flow between the two
species. Therefore, this estimate of 6 migrants per generation should be considered as a
maximum possible value for M.
When migration was incorporated into MIMAR, the prior values were sampled from
a range of ln(M) values between 0.135 and 7.39 (or roughly between M = 1 and M =
34
0
0.002
0.004
0.006
0.008
0.01
0.012
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35
Tcoal
T in coalescent units=Tgen/4N
1

(a) No Migration
0
0.002
0.004
0.006
0.008
0.01
0.012
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35
Tcoal
T in coalescent units=Tgen/4N
1

(b) With Migration
Figure 3.2: MIMAR estimates of the time since divergence between A. for-
mosa and A. pubescens. Both panels show the posterior distributions of the time
since divergence in coalescent time units. In the first panel, the distribution was gen-
erated by running MIMAR with the migration rate fixed at 0; in the second panel, the
prior distribution for the migration rate was U[0.135,7.39].
35
1600). Under this model, the estimate for the time since the split rose slightly from
557,845 to 660,860 years ago (Figure 3.2). Both of these estimates seem reasonable,
given that the diversification of the North American Aquilegia clade is believed to have
occurred less than 2 million years ago [31]. Despite this, however, MIMAR was not able
to converge on an estimate for the migration rate, even though the model seemed to be
mixingwellandtheestimatesofθ correspondedwelltoearliercalculationsinsection2.1.
This may be the result of having too little data in general, but it seems more likely that
itistheresultofhavingzerofixeddifferencesinthesample,especiallysinceotherstudies
of very recently diverged plants have encountered the same problems (Stephen Wright,
personal communication).
3.4 Summary
ThisimplementationoftheIMmodel[23,2]producedanestimateofthedivergencetime
that fit well with the model of recent speciation, but since it could not simultaneously
converge on an estimate for the migration rate, it is impossible to be sure that gene
flow is still occurring. Although these results are intriguing by themselves, they also
highlight the need for a full genome scan, which will provide more accurate estimates
of the proportions of species-specifc polymorphisms, shared polymorphisms, and fixed
differences, which in turn would result in more accurate MIMAR results.
To fully understand speciation, it is necessary to understand not just levels of gene
flow, but how migration and selection interact. If the migration rate exceeds the fitness
advantage of differences, then differentiation should be prevented [48]. But selection can
also drive the spread of advantageous alleles, even in the presence of migration. While
the identification of fixed differences and potential speciation genes will not definitively
prove or disprove a particular speciation model, comparing the pattern of variation in
these loci withthe patternofsharedvariationinneutralloci willprovide insightintothe
question of whether or not two species have diverged in the face of introgression [24, 3].
36
Chapter 4
Whole Genome Scan for
Speciation Genes
Ultimately, nostudyofspeciationcanbecompletewithoutfindingthelociunderpinning
species differences. Two of the questions central to the speciation process are: What
initiates the evolution of intrinsic reproductive isolation? and What causes this isolation
to increase until the speciation process is complete? When considering the issue from
a genomics standpoint, these questions become more focused: How many genes are
involved in the initiation of reproductive isolation? Can the same speciation scenario
happen in parallel in different locations? What types of genes and genetic variations are
typically responsible for adaptive differences between species? What is the strength of
selection on these genes? Does the rest of the genome become more differentiated over
time?
Next generation sequencing (NGS) allows for the rapid and cost-effective generation
of large amounts of sequence data for both model and non-model organisms alike, and
thus lends itself well to any kind of comparative genomics analysis. Several studies have
already availed themselves of this new technology to identify the genetic loci responsible
for adaptive evolution in a wide variety of systems including plants, fish, insects and
snails [63, 74, 37]. A natural extension of scanning for adaptive differences between two
populations is to scan for these same types of differences between two closely related
species [37], and the results from Chapters 2 and 3 have shown that Aquilegia formosa
and A. pubescens are excellent candidates for exactly this type of scan.
37
To identify the speciation loci and elucidate the mechanism of speciation of Aqui-
legia, whole genome sequences for A. formosa and A. pubescens were generated from
pooled population samples using the Illumina Genome Analyzer. Similarly to [74], the
aligned genomes were scanned for regions with significant disparity in allele frequency.
Because prior analyses of nuclear DNA sequences had so far revealed few high frequency
differences, it was known that this simple scan should not generate too many false posi-
tives, andthusallSNPsexhibitingextremedifferentiationcouldbeconsideredcandidate
speciation loci.
4.1 Alignment with Aquilegia coerulea
Oftheover800millionreadsobtained,approximately350million(44%)ofthemmapped
uniquely to the nuclear reference genome of Aquilegia coerulea (Goldsmith)(Table 4.1).
Thisresultedin26.4billionuniquelymappedbasepairs,whichis87timesaslargeasthe
301MbAquilegia genome. TheexpectedcoveragesforA. formosa andA. pubescens were
48.6X and 38.8X, respectively (Table 4.1). In actuality, the mode of coverage per species
was slightly lower than expected (closer to 30X instead of 40X)(Figure 4.1). Different
population pools also had varying levels of success, but this did not seem to be more
prevalent in one species versus another.
The discrepancy between the observed coverage and the expected coverage is most
likely the result of the repetitive nature of the Aquilegia genome. Many angiosperms
have a high repetitive element content in their genomes [72], and Aquilegia is no excep-
tion, with an estimated 41.1% of the reference genome being classified as repetitive by
RepeatMasker [62]. Although filtering for mapping quality should eliminate most reads
that align to repetitive regions, there is still a long tail observed in the distribution
shown in Figure 4.1, so that the average coverage value is equal to expected coverage,
even though the mode is not.
38
Table 4.1: Reads Mapping in Each Sample. Expected coverage values were calcu-
lated by dividing the number of base pairs in the Aquilegia genome (301 million) by the
total number of mapped base pairs.
Pool #Reads #Mapped Mapped bp Exp. Coverage
1 71,707,596 19,324,522 1,449,339,150 4.8
2 89,018,418 48,538,172 3,640,362,900 12.1
3 78,326,276 36,778,095 2,758,357,125 9.1
4 85,675,184 39,338,899 2,950,417,425 9.8
5 45,544,838 6,731,076 504,830,700 1.7
6 99,768,818 25,716,674 1,770,134,150 5.9
7 75,740,366 40,978,420 3,073,381,500 10.2
8 64,867,232 34,738,997 2,605,424,775 8.6
9 97,052,118 45,131,842 3,384,888,150 11.2
10 103,774,516 56,895,761 4,267,182,075 14.1
A. for. 467,324,430 195,842,606 14,688,195,450 48.6
A. pub. 344,150,932 158,329,852 11,716,122,500 38.8
Total 811,475,362 354,172,458 26,404,317,950 87.4
NotonlydoA. formosa andA. pubescens havesimilardistributionsofreadcoverage,
but the coverage at each base pair is significantly correlated between the two species
(Pearson correlation coefficient = 0.48, p = 9.5(10)
−12
). This is especially true for sites
withnocoverage, where4%ofbasepairshavenocoverageineitherspecies(comparedto
around6%whichhavenocoverageinatleastoneofthespecies)(Figure4.2). Giventhat
repetitive regionsinthe A. ceorulea genome willnotreceive manyuniquemappings, this
probably explains the majority of sites with no alignments in either species. A. formosa
and A. pubescens arealsomorecloselyrelatedtoeachotherthaneitheristo A. coerulea,
so it is not surprising that they exhibit highly similar patterns of failed and successful
mappings to the reference genome.
4.2 Genome-wide Levels of Polymorphism and Divergence
Wakeley and Hey [77] introduced four summary statistics that were useful to infer diver-
gence models from polymorphism data: the number of polymorphisms unique to each
39
0 50 100 150
0.00 0.01 0.02 0.03 0.04 0.05 0.06
Read Coverage per Base Pair
Proportion of Base Pairs
Figure 4.1: Read coverage per base pair for each species. The proportion of base
pairs is calculated using the total number of non-missing bases in the reference genome
(no Ns). The last point on the x–axis represents the combined data for read coverage
≥176. Red bars are values for A. formosa and blue bars are values for A. pubescens.
0 50 100 150 200
0.000 0.001 0.002 0.003 0.004 0.005
0
50
100
150
200
Coverage per base pair in Formosa
Coverage per base pair in Pubescens
Proportion of base pairs
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● Figure 4.2: Coverage correlation between the species. Values for read coverage
=0andcoverage=1arenotshownbecause the proportions aremuchhigherthanthose
on the rest of the plot. The proportions for these values are 4% and 1%, respectively.
40
species (S
1
and S
2
), the number of shared polymorphisms (S
s
) and the number of fixed
differences (S
f
). After filtering for coverage and quality criteria, there were a total
3,110,099 SNPs discovered: S
s
= 1,957,646; S
1
= 634,511; S
2
= 517,900; S
f
= 42.
Of the 1,957,646 shared polymorphisms, 97,286 (3% of all SNPs) were fixed differences
between A. coerulea and the other 2 species. Earlier, in section 2.1, it was found that
there were 2.5 times more shared than private polymorphisms; in this data set, that
ration is reduced to 1.7, but this still demonstrates that the majority of variation is
shared.
The nearly fixed differences fall onto 14 of the 971 nuclear genome scaffolds. In
several cases the region around the SNP seems to show a drop in the overall level of
shared polymorphism (Figure 4.3, see in particular Scaffold 3, position∼8Mb, the start
of scaffold 6, scaffold 31, position 1.5Mb, and scaffold 44, position 1Mb). There is
also a region on scaffold 31 which shows a conspicuous absence of data between two
differentiated sites. The region at the beginning of scaffold 8, which has the largest
number of nearly fixed sites, does not seem to show any particularly striking pattern of
shared or private polymorphism.
4.3 Coverage and its Confounding Effects
Because allele frequency and allele frequency differences between species are the main
parameters of interest in this study, it is crucial to ensure that they are being estimated
accurately from the Solexa data. If the coverage at a site is insufficient, then unfixed
variants are more likely to be mistakenly called fixed differences just by chance. For
instance, if the true SNP allele frequency in each species were 50%, and coverage was
only 1X for each species, then the site would either appear as non-polymorphic between
the species or as a fixed difference, and each of these scenarios would occur with equal
probability.
41
0e+00 2e+06 4e+06 6e+06 8e+06 1e+07
0.01 0.02 0.03 0.04 0.05 0.06
Scaffold 3
0e+00 2e+06 4e+06 6e+06 8e+06
0.00 0.01 0.02 0.03 0.04 0.05
Scaffold 4
0e+00 2e+06 4e+06 6e+06 8e+06
0.01 0.02 0.03 0.04 0.05
Scaffold 6
0e+00 2e+06 4e+06 6e+06
0.01 0.02 0.03 0.04 0.05
Scaffold 8
0e+00 1e+06 2e+06 3e+06 4e+06 5e+06 6e+06
0.01 0.02 0.03 0.04
Scaffold 13
0e+00 1e+06 2e+06 3e+06 4e+06 5e+06
0.01 0.02 0.03 0.04
Scaffold 15
0 500000 1000000 1500000 2000000 2500000 3000000 3500000
0.01 0.02 0.03 0.04
Scaffold 26
0 500000 1000000 1500000 2000000 2500000 3000000
0.01 0.02 0.03 0.04 0.05
Scaffold 31
0 500000 1000000 1500000 2000000
0.01 0.02 0.03 0.04
Scaffold 40
0 500000 1000000 1500000 2000000
0.01 0.02 0.03 0.04
Scaffold 44
0 500000 1000000 1500000 2000000
0.00 0.02 0.04 0.06 0.08
Scaffold 46
0e+00 1e+05 2e+05 3e+05 4e+05 5e+05 6e+05 7e+05
0.01 0.02 0.03 0.04 0.05 0.06 0.07
Scaffold 88
2e+05 3e+05 4e+05 5e+05 6e+05 7e+05 8e+05
0.01 0.02 0.03 0.04
Scaffold 90
0e+00 1e+05 2e+05 3e+05 4e+05
0.01 0.02 0.03 0.04 0.05 0.06
Scaffold 102
Figure 4.3: Sliding window analysis of polymorphism based on 10–kb windows
with no overlap. Grey lines indicated shared polymorphism, red lines are A. formosa
polymorphism, and blue lines are A. pubescens polymorphism. Green ticks along the
bottom of the graphs indicate positions of fixed or nearly fixed differences.
42
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● 0 20 40 60 80 100
0.00 0.02 0.04 0.06 0.08 0.10
Minimum within Species Coverage
Proportion of SNPs with Frequency Difference of at Least 90%
● Figure 4.4: Proportion of High Frequency Differences vs. Within Species
Coverage. At each within-species coverage value, the proportion of SNPs with an allele
frequency difference (AFD)≥90% was calculated out of the total number of SNPs with
that coverage. The red dot represents 30X coverage, which was determined to be the
best cut-off value.
A conservative coverage cut-off value will be the one where the proportion of high
frequency differences becomes constant even as coverage increases, a concept similar
to that of the false discovery rate described in [66]. Another way of stating this is
that for a given set of SNPs, the probability of finding a high frequency difference at
a single site should be statistically independent of the coverage at that site. If high
frequency differences are defined as SNPs with allele frequency differences (AFD)≥0.90,
then the proportion of SNPs above this threshold can be compared at each possible
minimum within-species coverage value (Figure 4.4). To systematically determine when
the proportion of high frequency SNPs was equal to the true proportion, a G-test for
independence was used [46] at each coverage threshold. In the case of AFD≥0.90, 24X
is the minimum coverage threshold. For thresholds of AFD as low as 60%, the ideal
coverage cut-off value tends to be at 30X, so this is the value that was ultimately used
(Figure 4.5).
43
Minimum within Species Coverage
AFD
Proportion of Sites
Figure 4.5: Allele Frequency Difference Threshold vs. Minimum Coverage.
Minimum coverage values range between 1 and 100X. The Allele Frequency Difference
(AFD) thresholds range between 5% and 100%, in increments of 5%.
4.4 Candidate Speciation SNPs
After establishing a conservative coverage cut-off value of 30X, it was possible to scan
the aligned genomes for SNPs with extreme differences in allele frequency. The top 1%
of SNPs have AFD≥60% (∼30,000 SNPs). If the search is limited to only fixed or nearly
fixed differences, then the number of SNPs is reduced to merely 42 (see also section 4.2).
Since the minimum AFD value for a nearly fixed SNP was∼93.5%, this was chosen as
the optimal threshold, and yielded 130 SNPs. These SNPs are distributed among a total
of 23 scaffolds, which together comprise∼31% of the genome.
44
Table 4.2: Highly Differentiated Non–Synonymous SNPs.
Scaffold Position AA Change Protein Function
3 7792786 Q→H AT3G58650.1 Unknown
15 599169 R→L MYB113 DNA binding,
transcription factor
31 1578156 T→I LPR2 copper ion binding,
oxidoreductase
31 2141877 A→T UGT78D2 UDP-Glucosyltransferase,
anthocyanidin
44 1102204 Q→H unknown unknown
47 1957601 L→P unknown unknown
In Figure 4.6, it can be seen that these extremely differentiated SNPs are often
located in “peaks” of differentiated SNPs (note that because of the scale of the plots in
thisfigure,peaksthatappeartobeaverticallineofdotsareactually∼10kbinsize). This
supportsthetheorythatdivergentselectionisactingonthesesites,sincelocaldistortions
in linkage disequilibrium are expected to occur around adaptively advantageous alleles
[22], and LD was shown to decay within 1kb in the previous sequencing survey of 9
neutral loci.
WhencomparingthegeneannotationsforthehighAFDSNPsversusallSNPsabove
the coverage threshold, it seemed that the high AFD SNPs were enriched for cDNA
sequences and for nonsynonymous mutations in particular (Figure 4.7). In the entire
Aquilegia genome,11%ofbasepairsarelocatedincDNAsequences,andthispercentage
increases to 18.7% when considering only non-repetitive DNA. In the full set of SNPs,
the percentage of SNPs annotated as being within a transcript matches the expected
percentage exactly (18.7%), although the fact that there are more nonsynonymous SNPs
(5.1%) than synonymous SNPs (2.9%) is surprising. In the high AFD SNPs set, 25.8%
of SNPs are within cDNA, and 7.4% of SNPs are nonsynonymous mutations (Table 4.2),
suggesting that the high AFD data set could contain important causal mutations.
45
Figure 4.6: Spatial Pattern of Allele Frequency Differences. The p–value of
each SNP site was calculated based on the observed distribution of AFD in the whole
sample. Different colors within each panel indicate separate scaffolds, which have been
grouped together randomly; positions to the right of the first scaffold in a panel have
been adjusted so that scaffolds could be plotted side by side. The horizontal dotted line
marks the−log
10
p for an allele frequency difference of 93.5%.
46
Intergenic
Within 5kb
Synonymous
NonSynonymous
Intron
Other
(a) All SNPs
Intergenic
Within 5kb
Synonymous
NonSynonymous
Intron
Other
(b) AFD≥ 93.5% SNPs
Figure 4.7: Gene Annotation for Different Subsets of SNPs.
4.4.1 Anthocyanin Pathway
One of the most promising results of this scan is a peak located on scaffold 15, around
600kbfromthestart(Figure4.6). Thisregionwaspreviouslyidentifiedasoneof2major-
effects QTLs controlling flower color, and was subsequently found to contain a cluster
of MYB family transcription factor genes (Scott Hodges, personal communication). The
geneannotationrevealedonenonsynonymousSNPwithinMyb113,whichcausedasingle
amino acid change of arginine (R) to leucine (L).
Several MYB transcription factors, including MYB113, are known to regulate the
production of anthocyanins (pigments) in a number of plants, although it appears that
Myb factors may regulate different genes in the flavonoid pathway for different species
[47, 19] (Figure 4.8). As previously mentioned, flower color is a key trait for attracting
different pollinators in Aquilegia, so any gene in the anthocyanin pathway is a strong
candidate for a speciation locus. It is worth noting that the other flower color QTL
covered a region containing DFR, a gene downstream of MYB (and directly regulated
by MYB in Petunias). In the Solexa analysis, no significant SNPs were found in this
47
Figure4.8: TheBranchofthePhenylpropanoidBiosyntheticpathwayyielding
anthocyanins. Arrows indicate which genes are regulated by Myb transcription factors
in different species. Figure adapted from [19].
region (located on scaffold 10, which is not shown in any figures due to lack of significant
or interesting data). However, another nonsynonymous change was found on scaffold
31, in a UGT gene that in soybeans was shown to encode the final step in anthocyanin
biosynthesis [36].
4.4.2 Inversions and Speciation
Another intriguing finding in these data is the large, highly differentiated region on
scaffold 8, which contains 49 SNPs with AFD≥93.5% (in addition to many other SNPs
at lower but still elevated levels of differentiation). Despite the striking pattern at this
locus, the annotation indicates that only 4 SNPs align to cDNA sequence, and only 2 of
these are in exons (but both are also synonymous changes). One of them is found in an
unknown protein, and the other is found in a transducin family protein/WD–40 repeat
48
family protein. These WD–40 repeats are found in almost all eukaryotes, and are known
to have diverse regulatory functions, so it is not immediately obvious what the role of
this protein might be in A. formosa or A. pubescens.
The differentiated SNPs on scaffold 8 seem to be interspersed among large sections
of repetitive sequence. Repetitive elements can, among other things, lead to genomic
rearrangements such as inversions, which have been found to be locally adapted in other
studies of speciation [41, 32]. Inversions are an important concept in speciation because
they suppress recombination, so even in a hybridizing species such as Aquilegia, sets of
coadapted genes situated in an inverted region will remain linked despite introgression
in the rest of the genome [41, 32]. Closely related species have often been found to differ
by chromosomal rearrangements, and these rearrangements often appear linked to eco-
logical adaptations (reviewed in [28]). While the presence of repetitive DNA and highly
differentiated polymorphisms is clearly not enough evidence to declare the presence of
an inversion on scaffold 8, it does indicate a need for further study of this region.
4.5 Materials and Methods
4.5.1 Sample Collection, Preparation, and Sequencing
Leaf and floral tissue were collected from a total of 10 different Aquilegia populations: 6
A. formosa and 4 A. pubescens (Table 4.3). All populations are located in the Eastern
Sierra Nevada region of California (see map in Figure 4.9).
Individual tissue samples were pooled by population, and DNA extractions were
performed using Qiagen’s DNeasy Plant Mini Extraction Kits, and concentrations were
assessedusingInvitrogen’sQubitfluorometer. ShearingofthegenomicDNAinto200-600
base-pair fragments was done using sonication. Briefly, the sonication protocol entailed
4 rounds of 20 pulses each (with 10 seconds on ice between rounds of pulses) with a duty
cycle of 80% and an output control of 1.8. A single sequence library was constructed
49
Table 4.3: Samples Used for Solexa Sequencing
Pool Species Population Tissue #Ind.
1 A. for. Lundy Canyon Lower (LCL1) floral 17
2 A. for. Rock Creek/Mosquito Flat (RC/MP2–MF) leaf 49
3 A. for. North Lake Stream (NLS) leaf 68
4 A. for. Bishop Pass Trail/South Lake (BPT/AR) leaf 37
5 A. for. Golden Trout Trail 1 (GT1) floral 41
6 A. pub. Morgan Pass Lower, Upper (MPL, MPU) floral 78
7 A. pub. Piute Pass (PT1) leaf 40
8 A. pub. Bishop Pass (BP) leaf 47
9 A. for. Gabel Lakes (GL) leaf 24
10 A. pub. Big Horn Lake (BHL) floral 35
Figure 4.9: Map of Sampling Locations. Red points represent A. formosa popula-
tions and yellow points represent A. pubescens.
50
for each pool according to standard Illumina protocols at the Gregor Mendel Institute
(GMI) in Vienna, Austria.
Each DNA library was sequenced using a paired-end protocol with 2 lanes of a flow
cell on an Illumina/Solexa Genome Analyzer at the GMI. An additional lane of Pool
#6 was also sequenced using a paired-end protocol at the USC Epigenomics Center in
Los Angeles, CA; this was initially done as a test for the method, and the results were
combinedwiththeresultsfromGMI.AllreadsfromtheGMIwere75bpinlength, while
readsfromUSCwere35bp. Intheend,atotalof811,475,362readswereobtained,which
translated into 60,053,208,390 base pairs of sequence data.
4.5.2 Read alignment and SNP calling
Sequencing reads were aligned to a draft assembly of the Aquilegia coerulea (Goldsmith)
genome(JointGenomeInstitute(JGI)Aquilegia SequencingProject,unpublisheddata).
The nuclear draft genome currently consists of 971 scaffolds, with additional scaffolds
for the plastid genomes. Mapping was performed using the Burrows-Wheeler Alignment
tool (BWA) [39] in paired-end mode with an allowance of up to 10 mismatches. Any
read with a mapping quality of zero was immediately discarded, since this is indicative
of non-unique alignment.
Becausethecoverageofindividualpopulationpoolswasdeemedinsufficientforaccu-
ratelyassessingallelefrequencies, dataweremergedforeachspeciesandPCRduplicates
wereremovedusingPicard[52]. Subsequently,SNPswerecalledusingSAMtools[40]and
adjusting the expected number of haplotypes based on the number of individuals known
to be in each population pool. Genotyping at each SNP site was done using in-house
scripts, and further filtering was performed to remove bases with a quality score <10.
Allele frequency was calculated as simply the fraction of reads with the SNP allele out
of the total number of reads covering a given site. Nearly fixed differences were defined
as sites where the removal of one read base in either species would result in an allele
frequency difference of 1.
51
4.5.3 Sliding Window Analyses
Scaffolds were divided into windows of 10kb with no overlap. A coverage cut-off was set
at a minimum of 30X within each species, and the number of bases within each window
meeting this requirement was determined. If this number was <1000, the window was
discarded from the analysis. The total number of SNPs in each window was also filtered
with the same coverage criterion, and the remaining SNPs were characterized as shared
or species–specific based on the derived allele frequency in each species. Only the data
for scaffold known to have a fixed or nearly fixed difference were plotted.
4.5.4 G–Test for Independence
Ateachminimumwithinspeciescoveragevalue,theproportionofSNPsaboveaspecified
AlleleFrequencyDifferentiation(AFD)thresholdwascalculatedforthatcoveragevalue,
and then the same proportion was calculated for all sites above that coverage value.
This was done for all AFD thresholds between 0.05 and 1 (in increments of 5%). The
two proportions were compared using the G–test for independence as described in [46],
provided that the expected values were not less than 5. If the expected values were
too small for the G–test, a Fisher’s Exact test was used instead. The proportion was
assumed to be independent of coverage once the test returned a p-value >0.05.
4.5.5 Annotation
Gene feature files for A. coerulea were downloaded from Phytozome [1], and SNPs were
annotatedusingsnpEff[5]. SinceitispossibleforaSNPtohavemultipleeffects,onlythe
most extreme effects were considered for analyses. For SNPs with high allele frequency
differences,itwasfoundthatalargeproportionofSNPscalledasintergenicwerelocated
inthesameregiononscaffold8, sothesesiteswereremovedtoavoidanybiastheymight
be introducing.
52
4.6 Summary
High-throughput sequencing enabled a successful and efficient scan for speciation genes
in Aquilegia formosa and A. pubescens, despite the limitations caused by a partially
assembled reference genome and highly repetitive DNA content. As predicted by the
initial sequencing survey of 9 nuclear genes, the majority of polymorphism was shared
between the two species, and selecting sites based on a threshold of allele frequency
difference produced a manageable number of candidate speciation loci. Two of these
genesalreadyhavewell-characterizedrolesintheanthocyaninbiosynthesispathwaythat
leadstofloralpigmentation,andgenotypingofSNPs(comingsoon!) inF2hybridsshould
elucidate the role of the other high frequency mutations.
The“peakiness”ofthedataaroundthefixedandnearlyfixeddifferencessupportsthe
theoretical expectations of differential introgression at sites experiencing divergent selec-
tion, and the set of SNPs produced by this project should prove very informative for any
future isolation–migration modeling and studies of sympatric vs. allopatric speciation.
53
Chapter 5
Conclusions
The goal of this project has always been to understand what causes and maintains
species differences between Aquilegia formosa and the closely related A. pubescens, and
the results presented in this thesis reflect a significant advancement towards this end. A
relatively small sequencing survey of the nuclear genome proved to be very informative,
andmostnotablyproducedtheintriguingresultthateventhoughtwospeciesareclearly
different in their morphologies and ecologies, they can be indistinguishable at the DNA
level across much of the genome. This finding, along with the ability of A. formosa and
A. pubescens to hybridize, challenges classical species concepts, but adds to a growing
number of recent studies that are rendering the traditional definitions of species and
speciation obsolete.
Regardless of how a “species” is defined, or whether or not a particular method of
speciationcanbeproved,A.formosa andA.pubescens representtwoentitieswithvisible,
heritableadaptivedifferencesbetweenthem,andthegeneticnatureofthesedifferencesis
ofgreatinterestfromanevolutionarybiologyperspective. BecausetheSangersequencing
results corresponded to what is predicted by population genetic theories of non-uniform
introgression in diverging taxa, it could be inferred that a whole-genome scan for fixed
differences would find a tractable number of candidates without being inundated by
false positives produced by genetic drift. This assumption proved to be correct, as the
high-throughput sequencing scan successfully pinpointed on the order of 100 mutations
that might be involved in reproductive isolation, and many of these appear to exhibit
signatures of natural selection.
The advent of high-throughput sequencing has undeniably brought about a break-
through in the study of adaptation, but thus far most whole genome comparisons have
54
been on different populations of the same species. Hopefully, the success of the Aquilegia
project will convince more evolutionary biologists that this simple and powerful method
can also be effective for two species, and that the newest technology can be used to
answer one of the oldest questions in biology: the origin of species.
55
Bibliography
[1] Phytozomeannotationfiles.ftp://ftp.jgi–psf.org/pub/JGI data/phytozome/v6.0/Acoerulea,
2010.
[2] Celine Becquet and Molly Przeworski. A new approach to estimate parameters of
speciation models with application to apes. Genome Research, 17:1505–1519, 2007.
[3] Roger K. Butlin, Juan Galindo, and John W. Grahame. Sympatric, parapatric or
allopatric: the most important way to classify speciation? Philosophical Transac-
tions of the Royal Society B: Biological Sciences, 363:2997–3007, June 2008.
[4] V.C. Chase and P.H. Raven. Evolutionary and ecological relationships between
Aquilegia formosa and A.pubescens (Ranunculaceae), two perennial plants. Evolu-
tion, 29:474–486, 1975.
[5] Pablo Cingolani. snpEff. http://snpeff.sourceforge.net/, 2010.
[6] Elizabeth A. Cooper, Justen B. Whittall, Scott A. Hodges, and Magnus Nordborg.
Genetic variation at nuclear loci fails to distinguish two morphologically distinct
species of aquilegia. PLoS One, 5(1):e8655, January 2010.
[7] Jerry A. Coyne and H. Allen Orr. Speciation. Sinauer Associates, Inc., Sunderland,
MA U.S.A., 2004.
[8] Karen D. Crow, Hiroyuki Munehara, Ziyusei Kanamoto, Andrey Balanov, Dmitriy
Antonenko,andGiacomoBernardi. Maintenanceofspeciesboundariesdespiteram-
pant hybridization between three species of reef fishes (Hexagrammidae): implica-
tions for the role of selection. Biological Journal of the Linnean Society, 91(1):135–
147, May 2007.
[9] Asher D. Cutter, Scott E. Baird, and Deborah Charlesworth. High nucleotide
polymorphism and rapid decay of linkage disequilibrium in wild populations of
Caenorhabditis remanei. Genetics, 174:901–913, October 2006.
[10] Charles Darwin. Narrative of the surveying voyages of His Majesty’s Ships Adven-
ture and Beagle between the years 1826 and 1836, describing their examination of
the southern shores of South America, and the Beagle’s circumnavigation of the
globe. Journal and remarks., volume 3. Henry Colburn, London, 1839.
56
[11] Charles Darwin. On the Origin of Species by Means of Natural selection or the
Preservation of Favored Races in the Struggle for Life. J. Murray, London, 1859.
[12] S. Dray and A.B. Dufour. The ade4 package: implementing the duality diagram for
ecologists. Journal of Statistical Software, 22(4):1–20, 2007.
[13] Brent Ewing and Phil Green. Base-calling of automated sequencer traces using
phred. II. Error probabilities. Genome Research, 8(3):186–194, March 1998.
[14] Brent Ewing, LaDeana Hillier, Michael C. Wendl, and Phil Green. Base-calling of
automatedsequencertracesusingphred.I.Accuracyassessment. Genome Research,
8(3):175–185, March 1998.
[15] J.Felsenstein.PHYLIP-phylogenyinferencepackage(version3.2).Cladistics,5:164–
166, 1989.
[16] Scott Freeman and Jon C. Herron. Evolutionay Analysis. Prentice-Hall, Upper
Saddle River, NJ 07458, 2 edition, 2001.
[17] M. Fulton and S.A. Hodges. Floral isolation between Aquilegia formosa and
Aquilegia pubescens. Proceedings of the Royal Society B: Biological Sciences,
266(1435):2247–2252, November 1999.
[18] S. Gavrilets. Fitness landsccapes and the origin of species. Princeton University
Press, Princeton, NJ, 2004.
[19] Antonio Gonzalez, Mingzhe Zhao, John M. Leavitt, and Alan M. LLoyd. Regula-
tion of the anthocyanin biosynthetic pathway by the ttg1/bhlh/myb transcriptional
complex in arabidosis seedlings. The Plant Journal, 53:814–827, 2008.
[20] David Gordon, Chris Abajian, and Phil Green. Consed: A graphical tool for
sequence finishing. Genome Research, 8(3):195–202, March 1998.
[21] Verne Grant. Isolation and hybridization between Aquilegia formosa and A.
pubescens. Aliso, 2(1):341–360, 1952.
[22] Daniel L. Hartl and Andrew G. Clark. Principles of Population Genetics. Sinauer
Associates, Inc., Sunderland, MA U.S.A., 4 edition, 2007.
[23] J. Hey and R. Nielsen. Multilocus methods for estimating population sizes, migra-
tion rates and divergence time, with applications to the divergence of Drosophila
pseudoobscura and D. persimilis. Genetics, 167:747–760, 2004.
[24] Jody Hey. Recent advances in assessing gene flow between diverging populations
and species. Current Opinion Genetics and Development, 16(6):592–596, December
2006.
[25] S.A. Hodges and M.L. Arnold. Columbines: A geographically widespread species
flock. Proceedings of the National Academy of Sciences, 91(11):5129–5132, May
1994.
57
[26] S.A. Hodges and M.L. Arnold. Floral and ecological isolation between Aquilegia
formosaandAquilegiapubescens. Proceedings of the National Academy of Sciences,
91(7):2493–2496, March 1994.
[27] Scott A. Hodges, Michelle Fulton, Ji Y. Yang, and Justen B. Whittall. Verne grant
and evolutionary studies of Aquilegia. New Phytologist, 161(1):113–120, January
2004.
[28] A.A.HoffmannandL.H.Rieseberg. Revisitingtheimpactofinversionsinevolution:
frompopulationgeneticmarkerstodriversofadaptiveshiftsandspeciation? Annual
Review of Ecological Evolution, 39:21–42, 2008.
[29] Richard R. Hudson, Montgomery Slatkin, and Wayne P. Maddison. Estimation of
levels of gene flow from DNA— sequence data. Genetics, 132(2):583–589, October
1992.
[30] T.H. Jukes and C.R. Cantor. Evolution of protein molecules, chapter p. 21-123.
Academic Press, New York, 1969.
[31] Kathleen M. Kay, Justen B. Whittall, and Scott A. Hodges. A survey of nuclear
ribosomal internal transcribed spacer substitution rates across angiosperms: an
approximate molecular clock with life history effects. BMC Evolutionary Biology,
6(36), April 2006.
[32] Mark Kirkpatrick and Nick Barton. Chromosome inversions, local adaptation, and
speciation. Genetics, 173:419–434, 2006.
[33] David B. Kitts. Plato on kinds of animals. Biology and Philosophy, 2(3):315–328,
1987.
[34] Thomas D. Kocher. Adaptive evolution and explosive speciation: The Cichlid fish
model. Nature Reviews Genetics, 5:288–298, April 2004.
[35] Judith M. Kolkman, Simon T. Berry, Alberto J. Leon, Mary B. Slabaugh, Shunxue
Tang, Wenxiang Gao, David K. Shintani, John M. Burke, and Steven J. Knapp.
Single nucleotide polymorphisms and linkage disequilibrium in sunflower. Genetics,
177:457–468, September 2007.
[36] Nik Kovinich, Ammar Saleem, John T. Arnason, and Brian Miki. Functional char-
acterization of a udp-glucose:flavonoid 3-o-glucosyltransferase from the seed coat of
black soybean (glycine max (l.) merr.). Phytochemistry, 71:1253–1263, 2010.
[37] M.K.N. Lawniczak, S.J. Emrich, A.K. Holloway, A.P. Regier, M. Olson, B. White,
S. Redmond, L. Fulton, E. Appelbaum, J. Godfrey, C. Farmer, A. Chinwalla, S.-P.
Yang,P.Minx,J.Nelson,K.Kyung,B.P.Walenz,E.Garcia-Hernandez,M.Aguiar,
L.D. Viswanathan, Y.-H. Rogers, R.L. Strausberg, C.A. Saski, D. Lawson, F.H.
Collins, F.C. Kafatos, G.K. Christophides, S.W. Clifton, E.F. Kirkness, and N.J.
Besansky. Widespread divergence between inicipient anopheles gambiae species
revealed by whole genome sequences. Science, 330:512–514, October 2010.
58
[38] A.Lawton-Rauh, R.H.Robichaux, andM.D.Purugganan. Diversityanddivergence
patternsinregulatorygenessuggestdifferentialgeneflowinrecentlyderivedspecies
of the Hawaiian silversword alliance adaptive radiation (Asteraceae). Molecular
Ecology, 16:3995–4013, 2007.
[39] H. Li and R. Durbin. Fast and accurate short read alignment with burrows-wheeler
transform. Bioinformatics, 25:1754–1760, 2009.
[40] H. Li, B. Handsaker, A. Wysoker, T. Fennell, J. Ruan, N. Homer, G. Marth,
G. Abecasis, R. Durbin, and 1000 Genome Project Data Processing Subgroup. The
sequencealignment/map(sam)formatandsamtools. Bioinformatics,25:2078–2079,
2009.
[41] David B. Lowry and John H. Willis. A widespread chromosomal inversion polymor-
phismcontributestoamajorlife-historytransition, localadaptation, andreproduc-
tive isolation. PLoS Biology, 8(9):e1000500, September 2010.
[42] Martin Maechler, Peter Rousseeuw, Anja Struyf, and Mia Hubert. Cluster analysis
basics and extensions, 2005. Rousseeuw et. al. provided the S original, which has
beenportedtoRbyKurtHornik,andhassincebeenenhancedbyMartinMaechler:
speed improvements, silhouette() functionality, bug fixes, etc. See the ’Changelog’
file (in the package source).
[43] James Mallet. Hybridization, ecological races and the nature of species: empirical
evidence for the ease of speciation. Philosophical Transactions of the Royal Society
B: Biological Sciences, 363:2971–2986, June 2008.
[44] N.Mantel.Thedetectionofdiseaseclusteringandageneralizedregressionapproach.
Cancer Research, 27:209–220, 1967.
[45] Ernst Mayr. Systematics and the Origin of Species. Columbia University Press,
New York, 1942.
[46] J.H. McDonald. Handbook of Biological Statistics. Sparky House Publishing, Balti-
more, MD, 2 edition, 2009.
[47] Frank Mehrtens, Harald Kranz, Pawel Bednarek, and Bernd Weisshaar. The
arabidopsis transcription factor myb12 is a flavonol-specific regulator of phenyl-
propanoid biosynthesis. Plant Physiology, 1238:1083–1096, June 2005.
[48] Carrie L. Morjan and Loren H. Rieseberg. How species evolve collectively: impli-
cations of gene flow and selection for the spread of advantageous alleles. Molecular
Ecology, 13:1341–1356, 2004.
[49] P.A. Munz. Aquilegia: the wild and cultivated columbines. Gentes Herbarum,
7:1–150, 1946.
[50] MohamedA.F.NoorandJeffreyL.Feder. Speciationgenetics: evolvingapproaches.
Nature Reviews Genetics, 7(11):851–861, November 2006.
59
[51] Magnus Nordborg, Tina T. Hu, Yoko Ishino, Jinal Jhaveri, Christopher Toomajian,
Honggang Zheng, Erica Bakker, Peter Calabrese, Jean Gladstone, Rana Goyal,
Mattias Jakobsson, Sung Kim, Yuri Morozov, Badri Padhukasahasram, Vincent
Plagnol, Noah A. Rosenberg, Chitiksha Shah, Jeffrey D. Wall, Jue Wang, Keyan
Zhao, Theodore Kalbfleisch, Vincent Schulz, Martin Kreitman, and Joy Bergelson.
The pattern of polymorphism in Arabidopsis thaliana. PLoS Biology, 3(7):1289–
1299, July 2005.
[52] Picard. http://picard.sourceforge.net, 2009.
[53] W. Prazmo. Cytogenetic studies on the genus Aquilegia. IV. fertility relationships
among the Aquilegia species. Acta. Soc. Bot. Poloniae, 34:667–685, 1965.
[54] Jonathan K. Pritchard, Matthew Stephens, and Peter Donnelly. Inference of pop-
ulation structure using multilocus genotype data. Genetics, 155(2):945–959, June
2000.
[55] Susan E. Ptak, David A. HInds, Kathrin Koehler, Birgit Nickel, Nila Patil, Den-
nis G. Ballinger, Molly Przeworski, Kelly A. Frazer, and Svante Paabo. Fine-scale
recombination patterns differ between chimpanzees and humans. Nature Genetics,
37(4):429–434, February 2005.
[56] Noah A. Rosenberg. Distruct: A program for the graphical display of structure
results. http:// rosenberglab.bioinformatics.med.umich.edu/distruct.html, 2002.
[57] Noah A. Rosenberg, Saurabh Mahajan, Sohini Ramachandran, Chengfeng Zhao,
Jonathan K. Pritchard, and Marcus W. Feldman. Clines, clusters, and the effect
of study design on the inference of human population structure. PLoS Genetics,
1(6):660–671, December 2005.
[58] Dolph Schluter. Ecological causes of adaptive radiation. American Naturalist,
148(s1):s40–s64, November 1996.
[59] Caroline Scotti-Saintagne, Stephanie Mariette, Ilga Porth, Pablo G. Goicoechea,
TeresaBarreneche, CatherineBodenes, KornelBurg, andAntoineKremer. Genome
scanning for interspecific differentiation between two closely related oak species
[QuercusroburL.andQ.petrea(Matt.)Liebl.]. Genetics,168:1615–1626,November
2004.
[60] Ole Seehausen. Hybridization and adaptive radiation. Trends in Ecology and Evo-
lution, 19(4):198–207, April 2004.
[61] George G. Simpson. Principles of Animal Taxonomy. Columbia University Press,
New York, 1961.
[62] A.F.A. Smit, R. Hubley, and P. Green. Repeatmasker open-3.0, 1996-2010.
60
[63] Jessica Stapley, Julia Reger, Philine G.D. Feulner, Carole Smadja, Juan Galindo,
RobertEkblom, ClairBennison, AlexanderD.Ball, AndrewP.Beckerman, andJon
Slate. Adaptation genomics: the next generation. Trends in Ecology and Evolution,
25(12):705–712, December 2010.
[64] Matthew Stephens and Peter Donnelly. A comparison of bayesian methods for hap-
lotype reconstruction from population genotype data. American Journal of Human
Genetics, 73(5):1162–1169, November 2003.
[65] MatthewStephens,NicholasJ.Smith,andPeterDonnelly. Anewstatisticalmethod
for haplotype reconstruction from population data. American Journal of Human
Genetics, 68(4):978–989, April 2001.
[66] JohnD.StoreyandRobertTibshirani. Statisticalsignificanceforgenomewidestud-
ies. Proceedings of the National Academy of Sciences, 100(16):9440–9445, August
2003.
[67] Jared L. Strasburg and Loren H. Rieseberg. Molecular demographic history of the
annual sunflowers Helianthus annuus and H. petiolaris-large effective population
sizes and rates of long-term gene flow. Evolution, 62(8):1936–1950, August 2008.
[68] Fumio Tajima. Evolutionary relationship of DNA sequences in finite populations.
Genetics, 105(2):437–460, October 1983.
[69] R.J.Taylor. Interspecifichybridizationanditsevolutionarysignificanceinthegenus
Aquilegia. Brittonia, 19:379–390, 1967.
[70] R Development Core Team. R: A Language and Environment for Statistical Com-
puting. R Foundation for Statistical Computing, Vienna, Austria, 2007.
[71] A.R. Templeton. The meaning of species and speciation: a genetic perspective.
In D. Otte and J.A. Endler, editors, Speciation and its Consequences, pages 3–27.
Sinauer Associates, Inc., Sunderland, MA U.S.A., 1989.
[72] Maud I. Tenaillon, Jesse D. Hollister, and Brandon S. Gaut. A triptych of the
evolution of plant transposable elements. Trends in Plant Science, 15:471–478,
2010.
[73] M.I. Tenaillon, M.C. Sawkins, A.D. Long, R.L. Gaut, J.F. Doebley, and B.S. Gaut.
Patterns of DNA sequence polymorphism along chromosome I of maize (Zea mays
ssp. mays L.). Proceedings of the National Academy of Sciences, 98(16):9161–9166,
2001.
[74] Thomas L. Turner, Elizabeth C. Bourne, Eric J. Von Wettberg, Tina T. Hu, and
Sergey V. Nuzdhin. Population resequencing reveals local adaptation of arabidopsis
lyrata to serpentine soils. Nature Genetics, 42(3):260–263, March 2010.
[75] Thomas L. Turner, Matthew W. Hahn, and Sergey V. Nuzdhin. Genomic islands of
speciation in Anopheles gambiae. PLoS Biology, 3(9):1572–1578, September 2005.
61
[76] Sara Via. Sympatric speciation in animals: the ugly duckling grows up. Trends in
Ecology and Evolution, 16(7):381–390, July 2001.
[77] J. Wakeley and J. Hey. Estimating ancestral population parameters. Genetics,
145:847–855, 1997.
[78] G. A. Watterson. On the number of segregating sites in genetical models without
recombination. Theoretical Population Biology, 7(2):256–276, April 1975.
[79] B.S. Weir and W.G. Hill. Nonuniform recombination with the human beta-globin
gene cluster. American Journal of Human Genetics, 38:776–778, 1986.
[80] Justen B. Whittall and Scott A. Hodges. Pollinator shifts drive increasingly long
nectar spurs in columbine flowers. Nature, 447:706–709, June 2007.
[81] Justen B. Whittall, Andrew Medina-Marino, Elizabeth A. Zimmer, and Scott A.
Hodges. Generating single-copy nuclear gene data for a recent adaptive radiation.
Molecular Phylogenetics and Evolution, 39(1):124–134, April 2006.
[82] Kenneth H. Wolfe, Wen-Hsiung Li, and Paul M. Sharp. Rates of nucleotide sub-
stitution vary greatly among plant mitochondrial, chloroplast, and nuclear DNAs.
Proceedings of the National Academy of Sciences, 84:9054–9058, December 1987.
[83] S. Wright. The genetical structure of populations. Annals of Eugenics, 15:323–354,
1951.
[84] Ji Y. Yang, Brian A. Counterman, Christopher G. Eckert, and Scott A. Hodges.
Cross-species amplification of microsatellite loci in Aquilegia and Semiaquilegia
(Ranunculaceae). Molecular Ecology Notes, 5(2):317–320, 2005.
[85] Yoko Yatabe, Nolan C. Kane, Caroline Scotti-Saintagne, and Loren H. Rieseberg.
Rampant gene exchange across a strong reproductive barrier between the annual
sunflowers, Helianthus annuus and H. petiolaris. Genetics, 175(4):1883–1893, April
2007.
[86] Y.L.Zhu,Q.J.Song,D.L.Hyten,C.P.VanTassell,L.K.Matukumalli,D.R.Grimm,
S.M. Hyatt, E.W. Fickus, N.D. Young, and P.B. Cregan. Single-nucleotide poly-
morphisms in soybean. Genetics, 163:1123–1134, March 2003.
62

Abstract (if available)

Abstract Aquilegia formosa and Aquilegia pubescens are two closely-related species belonging to the columbine genus. Despite their morphological and ecological differences, previous studies have revealed a large degree of intercompatibility as well as little sequence divergence between these two taxa, and the genetic mechanisms underpinning reproductive isolation remain unknown. In order to assess the feasibility of a full genome scan for speciation genes, inter- and intraspecific patterns of variation were compared for 9 nuclear loci

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Analysis of genomic polymorphism in Arabidopsis thaliana

PDF

Genome sequencing and transcriptome analysis of the phenotypically plastic spadefoot toads

PDF

Cellular level bottlenecks: genetic diversity, population dynamics, and technology development

PDF

Two-stage genotyping design and population stratification in case-control association studies

PDF

Biological interactions on the behavioral, genomic, and ecological scale: investigating patterns in Drosophila melanogaster of the southeast United States and Caribbean islands

PDF

From gamete to genome: evolutionary consequences of sexual conflict in house mice

PDF

Long term evolution of gene duplicates in arabidopsis polyploids

PDF

Efficient two-step testing approaches for detecting gene-environment interactions in genome-wide association studies, with an application to the Children’s Health Study

PDF

Exploring the genetic basis of quantitative traits

PDF

Mapping epigenetic and epistatic components of heritability in natural population

PDF

New tools for whole-genome analysis of DNA replication timing and fork elongation in saccharomyces cerevisiae

PDF

Evolutionary genomic analysis in heterogeneous populations of non-model and model organisms

PDF

Plant genome wide association studies and improvement of the linear mixed model by applying the weighted relationship matrix

PDF

Model selection methods for genome wide association studies and statistical analysis of RNA seq data

PDF

Studies in bivalve aquaculture: metallotoxicity, microbiome manipulations, and genomics & breeding programs with a focus on mutation rate

PDF

Decoding the embryo: on spatial and genomic tools to characterize gene regulatory networks in development

PDF

Bayesian hierarchical models in genetic association studies

PDF

Male-female conflict after mating: function and dynamics of the copulatory plug in mice (Mus domesticus)

Asset Metadata

Creator Cooper, Elizabeth Armstrong (author)

Core Title A population genomics approach to the study of speciation in flowering columbines

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Molecular Biology

Publication Date 02/09/2011

Defense Date 01/04/2011

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag adaptation,Aquilegia,Evolution,genomics,OAI-PMH Harvest,Solexa,speciation

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Nordborg, Magnus (committee chair), Conti, David V. (committee member), Finkel, Steven E. (committee member), Nuzhdin, Sergey (committee member), Wu, Xuelin (committee member)

Creator Email eacooper@usc.edu,lizcooper2@yahoo.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m3651

Unique identifier UC1201523

Identifier etd-Cooper-4331 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-427061 (legacy record id),usctheses-m3651 (legacy record id)

Legacy Identifier etd-Cooper-4331.pdf

Dmrecord 427061

Document Type Dissertation

Rights Cooper, Elizabeth Armstrong

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu