University of Southern California Dissertations and Theses

Geodatabase for archaeogenetics: ancient peoples and family lines

(USC Thesis Other)

Geodatabase for archaeogenetics: ancient peoples and family lines

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

Geodatabase for Archaeogenetics:

Ancient Peoples and Family Lines

by

Maria Leasure

A Thesis Presented to the
FACULTY OF THE USC DORNSIFE COLLEGE OF LETTERS, ARTS AND SCIENCES
University of Southern California
In Partial Fulfillment of the
Requirements for the Degree
MASTER OF SCIENCE
(GEOGRAPHIC INFORMATION SCIENCE AND TECHNOLOGY)

August 2021

Copyright © 2021 Maria Leasure

ii

Dedication

To Jason, Qing’an, and Valentino for your faith, love, and support.

iii

Acknowledgements
I would like to express my appreciation for conversations with Curtis Rogers concerning
the relatively new field of genetic genealogy, and the volunteer work done by Open Street Map
contributors in historical map projects.
I am indebted to Professor Laura Loyola for the insight and initial ideas to formulate this
topic. I equally owe deep thanks to Professor Wu An-min, for inspiring an interest beyond
“mapping” into the data and necessary structures and continuing to guide as a committee
member, and to Professor Chiang Yao-yi, for generously sacrificing time to serve on the
committee and share his expertise and important critiques. I am very grateful to Professor
Jennifer Bernstein, my advisor, for so many improvements, motivation, and strong moral
support.
The end of this project would not have come without a strong beginning, for which I
thank Professors Darren Ruddell and Vanessa Osborne, who helped me overcome anxieties and
direct my first efforts, and Robyn McNabb and Dr. Robert Vos for their belief and
encouragement. A special thanks to Professor “Kirk” Oda, who taught me to love technical
challenges and to learn from mistakes, and to Professor Elisabeth Sedano, who also helped me
become more adaptable and appreciate the curveballs. All these individuals have improved not
only a MS project but a person.
I am grateful to my colleagues, Samantha Bamberger and Timothy Williams, who helped
accommodate my schedule to focus on this work at critical times, and from whom I have learned
immeasurable amounts, and my classmates and friends at USC, who are unfailingly kind and
ready to help. For a sounding board on all phases of this work and support in every possible way,
I thank Jason Leasure.

iv

Table of Contents
Dedication .................................................................................................................................. ii
Acknowledgements ................................................................................................................... iii
List of Tables ............................................................................................................................vii
List of Figures ......................................................................................................................... viii
Abbreviations ............................................................................................................................. ix
Abstract ...................................................................................................................................... x
Chapter 1 Introduction ................................................................................................................ 1
1.1.1. Outline ................................................................................................................... 1
1.2. Objective ........................................................................................................................ 2
1.2.1. Geodatabase Design to Aid Spatial Analysis .......................................................... 2
1.2.2. Scope: Enabling Wider-Scale Studies..................................................................... 3
1.2.3. Advantages of Geodatabase Format ....................................................................... 3
1.3. Motivations .................................................................................................................... 5
1.3.1. Spatial Analysis Essential to Migration Studies ...................................................... 5
1.3.2. Spatial Database Precedent in Related Fields ......................................................... 5
1.4. Intended Users and Applications ..................................................................................... 6
1.4.1. Exploring Global Connections ............................................................................... 6
1.4.2. Multi-Disciplinary Work ........................................................................................ 7
1.4.3. Geography and Spatial Distributions ...................................................................... 8
1.5. Overview of Methodology .............................................................................................. 8
1.5.1. Overview ............................................................................................................... 9
Chapter 2 Related Disciplines and Literature Review ................................................................ 10
2.1. Genetics in Population Studies ...................................................................................... 10
2.1.1. The Primacy of Nonrecombinant Markers ............................................................ 12

v

2.1.2. Objections to Reliance on Non-Recombinant Markers ......................................... 13
2.2. Geodatabases as Aids to Analysis ................................................................................. 14
2.2.1. Geodatabases in Biology ...................................................................................... 14
2.2.2. The “Internet Museum” and Digital Collections ................................................... 15
2.2.3. Databases and Geodatabases in Archaeology ....................................................... 16
2.3. Developing the Geodatabase ......................................................................................... 18
2.3.1. Beginnings - Esri ................................................................................................. 18
2.3.2. Spatial Data Model .............................................................................................. 18
2.3.3. Not Limited to Proprietary Format ....................................................................... 19
Chapter 3 Data and Research Methods ...................................................................................... 20
3.1. Data Needs and Acquisition .......................................................................................... 20
3.1.1. Spatial Context .................................................................................................... 21
3.1.2. Finds – the Central Data Set of Ancient Human Specimens .................................. 21
3.1.3. Genetic Data – the Family Lines of mtDNA and Y-DNA ..................................... 24
3.2. Software and Tools ....................................................................................................... 25
3.2.1. Prototype and Final Software Solution ................................................................. 26
3.2.2. Software Solutions for Complete Spatial Database ............................................... 26
3.3. Database Design ........................................................................................................... 27
3.3.1. Overview of Tables .............................................................................................. 28
3.3.2. Relationships and Cardinalities ............................................................................ 29
3.3.3. New vs. Original Design for Tables ..................................................................... 30
3.3.4. Data Types ........................................................................................................... 30
3.3.5. Region Feature Class ........................................................................................... 31
3.3.6. Archaeological_find Feature Class ....................................................................... 32
3.3.7. Genetic Data Tables ............................................................................................. 34

vi

3.4. Testing the Adequacy of the Data and Functionality of the Database ............................ 35
Chapter 4 Results ...................................................................................................................... 37
4.1. Final Database Design .................................................................................................. 37
4.1.1. ERD Changes ...................................................................................................... 37
4.2. Moving Away from a Standard Relational Database ..................................................... 38
4.2.1. Final Schema as Implemented in ArcGIS Pro ....................................................... 39
4.2.2. Aspects of Relational Database Structure ............................................................. 40
4.3. Resulting Queries and Visualizations ............................................................................ 43
4.3.1. Query with Spatial Join Test – DNA Tables and Region ...................................... 43
4.3.2. Relationship Class Test – Mapping Related Data ................................................. 44
4.3.3. Test of Querying against Multiple Fields – Y-DNA Examples and Date Fields .... 46
4.3.4. Make Query Table – an Alternative to Standard Joins and Definition Queries ...... 48
Chapter 5 Conclusions and Discussion ...................................................................................... 53
5.1. Overview ...................................................................................................................... 53
5.2. Data Gaps ..................................................................................................................... 55
5.2.1. Acknowledging the Data Gap .............................................................................. 55
5.2.2. Technology and Global Sample Availability ........................................................ 56
5.2.3. Cultural, Legal, and Moral Considerations ........................................................... 57
5.3. Lessons Learned ........................................................................................................... 60
5.3.1. Domains .............................................................................................................. 60
5.3.2. Tool Use in the File Geodatabase ......................................................................... 61
5.4. Future Possibilities ....................................................................................................... 62
5.4.1. Open Source Options? ......................................................................................... 62
References ................................................................................................................................ 65
Appendix A Data Table ............................................................................................................. 69

vii

Appendix B Published Data ...................................................................................................... 70
Appendix C Domains ................................................................................................................ 72
List of Tables
Table 1 Original data spreadsheet .............................................................................................. 32
Table 2 Data types for ‘archaeological_find’ ............................................................................. 43
Table 3 The mtDNA genetic clades and upstream haplogroups ................................................. 45
Table 4 Make Query Table results for earliest Y-DNA haplogroups .......................................... 59
Table 5 Make Query Table restricted date results for ‘archaeological find’ using mtDNA ......... 60
Table 5 Make Query Table restricted date results for mtDNA haplogroup V ............................. 61

viii

List of Figures
Figure 1 Table view in ArcCatalog after feature class conversion .............................................. 34
Figure 2 Conceptual ERD ......................................................................................................... 38
Figure 3 Schema drawing with data types of early geodatabase plan ......................................... 41
Figure 4 Geography view of feature class ‘archaeological_find’ points in Catalog..................... 43
Figure 5 Conceptual ERD as implemented ................................................................................ 48
Figure 6 Schema drawing with data types for geodatabase ........................................................ 50
Figure 7 Create Relationship Tool and parameters..................................................................... 51
Figure 8 Relationship Classes in geodatabase as viewed in Catalog ........................................... 52
Figure 9 Relationship Class properties as seen in Catalog .......................................................... 52
Figure 10 Spatial join query, Q haplogroup ............................................................................... 54
Figure 11 Query using relationships, mtDNA haplogroup ........................................................ 55
Figure 12 R1a and R1b distributions before 1 CE ...................................................................... 57
Figure 13 Make Query Table Processing Parameters interface................................................... 59

ix

Abbreviations
BCE Before the Common Era
CE Common Era, equivalent to AD anno domini
DEM Digital Elevation Model
DNA Deoxyribonucleic acid
GIS Geographic information system
GISci Geographic information science
kya Thousands of years ago
mtDNA Non-recombining DNA from the mitochondria
SSI Spatial Sciences Institute
SNP Single Nucleotide Polymorphism
USC University of Southern California
Y-DNA Non-recombining DNA from the Y chromosome

x

Abstract
From its early beginnings from the parent genus Homo in Africa, the species Homo sapiens
spread across the globe to every continent except Antarctica, long before the advent of large
seafaring vessels or even the wheel. The dispersion of the first Homo sapiens occurred when
other early human species, such as Neanderthals or Denisovans, were still in Europe and parts of
Asia, and land features and climates were very different from northern and eastern Africa. As
early modern humans encountered these new environments and possibly other, earlier peoples
over centuries of migration, adaptations occurred, and new cultures arose. These migrations are
of great interest to several disciplines, including physical anthropology, archaeology, and
genetics. A global geodatabase as a repository of spatial and genetic data to facilitate Spatio-
temporal models of models and various data visualizations would serve all these disciplines.
Such a geodatabase also can incorporate other related data for investigation, such as global
regions, early coastlines, glacier limits, or the overall continental geography of earlier ages for
investigation of their possible effects on movement or settlement of ancient peoples.
Additionally, a geodatabase offers many options to share or limit access to data. This project
offers a comprehensive data source and tool for creating and sharing analyses with other research
efforts.

1

Chapter 1 Introduction
Many species remain forever restricted to a certain habitat essential to their survival, only
to be found in one region on the earth. Others, like our species Homo sapiens, succeed in
adapting to diverse conditions and expand their range across multiple continents. Humans
dispersed early in their history and developed many different material cultures in response to
these varied environments, as seen in tools, weapons, and other items discovered today. They
also developed great diversity in other aspects of culture, speaking numerous languages for
communication, establishing a variety of religious belief systems, and creating unique forms of
art. To understand the impact of these migrations and the rich diversity of our species, specialists
in biology, chemistry, genetics, anthropology, archaeology, and other fields study not only the
artefacts left by early peoples but their remains. Work specifically related to their genetic
markers – sometimes called archaeogenetics – is adding a new dimension to these studies.
1.1.1. Outline
This chapter first introduces key problems in studies of ancient peoples, their cultures,
and the impact of archaeogenetics on established ideas. In section 2, the chapter discusses the
objective of fulfilling unmet needs of consolidated data for research, which will provide for
essential spatial analysis (described in sub-section 1.2.1) and enabling wider scale studies (1.2.2).
The advantage of the geodatabase format to these goals is addressed in sub-section 1.2.3. Section
3 shares the initial motivations for building a database as a tool inspired by the visualization of
migrations (1.3.1) and by a successful application of a spatial database in genetics (1.3.2).
Section 4 expands the objective and motivations by describing the geodatabase’s applications
and intended users in more specific scenarios, including unexpected global connections (1.4.1),
increased multi-disciplinary work (1.4.2), and geographic distributions studies (1.4.3). Section 5

2

gives a basic methodology description and presents a more general outline and overview for the
project in 1.5.1.
1.2. Objective
This geodatabase will bridge the efforts of researchers in different fields investigating
early origins, migrations, and relationships of ancient peoples by offering a tool useful for multi-
disciplinary and global spatial analysis. Many works have been published on single excavation
sites or regional investigations, or geographically broader studies of specific time periods.
However, a comprehensive, consolidated resource for reference and analysis is lacking in a
format that can be readily used by software tools. The geodatabase is intended to be such a
resource, by providing a unified spatial catalogue of a variety of data collected from the
excavation and laboratory investigation of ancient human remains.
1.2.1. Geodatabase Design to Aid Spatial Analysis
A comprehensive spatial catalogue of data that can be queried will aid spatial analysis in
migration and cultural studies. For example, to what extent past languages and cultural practices
were spread through observations or exchanges between neighboring groups and how much was
a product of actual migration is still researched and debated by Cavalli-Sforza (1994), Renfrew
(2014), and others. Spatial analysis can determine which scenarios are more likely. However,
without a unified resource for the cultural and genetic data necessary, this analysis and
visualization is more difficult and time-consuming. A geodatabase is a valuable tool to provide
structure for an analytic process. This geodatabase will include some data concerning the cultural
identification of individuals whose DNA is catalogued here. The design of the geodatabase will
allow for the future incorporation of cultural data sets, such as extents of spoken languages.
Evidence of early genetics can support or cast doubt on theories in other studies of ancient

3

peoples, such as reconstructed migrations based upon linguistics. A genetic geodatabase provides
storage and structure to use spatial analysis with this evidence.
1.2.2. Scope: Enabling Wider-Scale Studies
This project aims to include archaeological finds dated from the oldest Paleolithic
discoveries to Iron Age peoples at the beginnings of recorded human history, from every
continent, to provide a technology-friendly resource that is not limited in scope to a single place,
time, or culture. The geodatabase also will include genetic marker data in addition to cultural
assessments, estimated dates, and the spatial data to map the archaeological finds.
Existing publications generally are restricted to one location or subdivision of an epoch to
allow the focus required for detailed study, such as that of a Later Stone Age burial site near
Eulau, Germany (Haak et al. 2008). In Eulau, the in-depth investigation of several members of a
family group revealed many insights into the lives of early peoples, including their social
structures and family practices, such as a pattern of men seeking partners outside the local area.
These localized, more granular studies are necessary to determine social patterns and interactions
of ancient peoples. However, although they are important as evidence to form and evaluate
theories on a wider scale, some effort to aggregate them must be made for this broader analysis.
This geodatabase project gathers data from many separate studies of this type and unifies them
for analysis on a larger scale, both in terms of time and place.
1.2.3. Advantages of Geodatabase Format
The geodatabase has several distinct advantages to researchers besides serving as a
consolidated source of data. The geodatabase is superior to a simple folder collection of geojsons
or kmls for example, because unlike files that use WGS84 (“Converting GIS Vector Data to
KML | Keyhole Markup Language” n.d.; “RFC 7946 - The GeoJSON Format” n.d.), the feature

4

classes can be reprojected into a coordinate system more suitable for smaller regional or local
analysis. The shapefile format shares this advantage, but again the geodatabase has advantages in
organization. A shapefile is not only a single .shp but a collection of files, including .dbf and
.shx, and if these components are missing, the shapefile is useless. The shapefile also could
include other files, such as .cpg or .prj, and a loss of these files may cause issues with
performance. When transferring a collection of shapefiles, care must be taken to not only move
the primary .shps, but all their other essential components. A geodatabase provides one item to
copy, move, or share over a network. Data stored and organized in a geodatabase potentially
could simplify data retrieval and preparation for use in other software tools besides GIS desktop
programs, thanks to its structure.
Many static or interactive maps with archaeogenetic themes are available, but unlike
these, a geodatabase makes use of aspects of relational databases to allow for queries, kernel
density estimation, and other operations beyond simple visual presentation. Because it can be
accessed by the most used GIS desktop tools, the geodatabase can be used to test new models of
migrations and spatial distributions of ancient groups. The aspects of relational databases can be
exploited to simplify the analytic process; by setting up initial permanent relationship classes,
repeated relate and join operations can be avoided.
These relationship classes can do more than temporary joins and relates. They also
establish a referential integrity, such as when attributes are changed then their related objects
may be automatically updated. The relationship class aids the edit process by granting quicker
access to related objects. The advantage of editing one entry in a related table in a geodatabase
over locating and updating multiple instances across several tables and joined shapefiles is
obvious, both for convenience and maintenance overhead as well as data integrity.

5

1.3. Motivations
Clues to the daily lives of ancient peoples are present not only through the evidence of
the objects they left behind but also their remains. The development of genetic analysis has
added another component to the investigation and analysis of early human remains. There are
several theories concerning the migrations of our earliest ancestors in Africa and their first
journeys to other continents (Beyin 2011). Despite new techniques in analysis and a growing
body of literature, no established global database of early human finds yet exists with the
required data for this multi-disciplinary spatial analysis.
1.3.1. Spatial Analysis Essential to Migration Studies
A study of any species must consider its environment; such a study without a sense of
place is incomplete. It naturally follows that these studies of early peoples would make extensive
use of spatial analysis and visualization to reconstruct the patterns of migration and settlement.
Beyin’s (2011) mapping of some of the earliest human migrations out of Africa is one such
reconstruction. Some research teams have used genetic markers of human remains and
comparisons with modern-day haplogroup distributions to find possible paths of migration not
previously considered (Reich et al. 2012; Achilli et al. 2013). A geodatabase of the data used by
these researchers would be useful to replicate their efforts and expand upon these ideas.
1.3.2. Spatial Database Precedent in Related Fields
The investigation of archaeogenetics, or the DNA characteristics of early peoples, is
important to many investigations of ancient groups and their migrations. One aspect of
archaeogenetics is the distribution of various haplogroups. Several efforts to map the occurrence
of these haplogroups geographically in modern populations have been successful. In 2005,
McDonald of the University of Illinois published maps showing the percentage of various DNA

6

haplogroups present today in different regions of the world. More recently, a database for the
present-day distribution of mtDNA, the non-recombinant DNA passed exclusively through the
maternal line, has been created (Rasheed et al. 2017) called “mtDNAmap”. It is a spatial
database implemented using MySQL. Access to the database is not limited to academics, as
Rasheed’s team has enabled access to the public via the Internet at URL
http://www.dnageography.com/mtDNAmap.php.
1.4. Intended Users and Applications
New research by specialists in genetics, anthropology, and anthropology challenges long-
standing theories on the arrival of modern humans and their arrival in certain geographic areas.
The dominant theory of human origins is that of an African origin, often assumed to be one
location from which early humans spread to the rest of the continent, but work in north Africa
suggests that from earliest times the species was found in more than one region (Hublin et al.
2017). Genetics work also has caused a re-evaluation of the traditional theory on human
settlement of the Americas by one wave of hunter-gatherers from Siberia by way of Beringia, a
former landmass now covered by the waters of the Bering Strait (Achilli et al. 2013; Reich et al.
2012; Wei et al. 2018; Yang et al. 2010). Many opportunities for continued inquiry in these and
other regions of the world remain and would benefit from a geodatabase to categorize, analyze,
and visualize data.
1.4.1. Exploring Global Connections
For many years, the early 1900s view of Aleš Hrdlička that early peoples from Siberia
came across a land bridge into the Americas in one large wave of migration (Hrdlička 1907;
Hrdlicka 1936) remained pre-eminent. Later researchers have found evidence to challenge this,
some even finding a three-wave view too simplistic (Arias et al. 2018; Achilli et al. 2013; Reich

7

et al. 2012). Teams led by Yang (2010) and Skoglund (2015) also found evidence for migrations
over water, rather than a slow overland trek, and some evidence of a link between South America
and Polynesia emerged in other genetic work by Gonçalves et al. (2013). A global geodatabase
of human samples and their genetic attributes could assist in the exploration of links between
widely separated incidences of genetic markers.
1.4.2. Multi-Disciplinary Work
Researchers have explored the movements of peoples in Europe and Asia using in-depth
analysis of genetics, while also considering archaeologists’ research of cultural artifacts and
linguists’ constructions of language trees. Prominent scholar (Gimbutas 1963) theorized that a
linguistic wave originating from a unified group (i.e. the Kurgan culture) crossed from the Pontic
steppes to southern and western regions of Europe. Cavalli-Sforza also created reconstructions of
the migrations of early peoples throughout Asia and Europe using the study of languages
alongside archaeogenetics (Cavalli-Sforza 1994). During the early years of archaeogenetics,
researchers Sokal, Oden, and Thomson (1992) countered that neither large scale Pontic steppes
migration nor other competing theories completely explained language distribution in Europe.
When geography was held constant, Sokal, Oden and Thomson (1992) found a “markedly
lower” correlation between language and genetics, which they admitted was still statistically
significant. However, later researchers with data from newer genetic extraction and analysis
techniques found evidence supporting the earlier ideas of a group of related, migrating people
carrying their language from the steppes into new territories (Krzewińska et al. 2018).
Investigation including geospatial analysis using new genetic data can either contradict or
support earlier theories derived from work in physical anthropology, linguistics, or other fields.

8

1.4.3. Geography and Spatial Distributions
Even though genetic studies can reconstruct trees of haplogroups and clades, the role of
geography in understanding pathways and motivations for migrations cannot be overstressed.
Yang et al. (2010) found that dispersal of genetic markers from the extreme northwestern parts of
the Americas correlated to a least-cost path using coasts as facilitators and looked forward to
future studies on the impact of other major geographic features such as mountains and river
valleys. Baumann (2017) used agent-based models in similar explorations of the impact of
coastline navigation in the Mediterranean. Baumann (2017) constructed these models to assess
the feasibility of using early seacraft to fish, hunt, or settle in islands during the earliest ages of
humanity – including times when earlier, related species as Homo neanderthalensis existed in
large numbers. Previously, scientists believed that such early peoples could only journey over
land given the level of technology and seafaring knowledge supposedly held at that time
(Baumann 2017). This project will include basic geographic data, but its geodatabase structure
can easily accommodate raster or vector files delineating ancient coastlines, glacial expanses,
and other potential factors that facilitate or impede migrations.
1.5. Overview of Methodology
The organization of data from these finds into a geodatabase could aid anthropologists
and genetic researchers in several ways. Users can identify each archaeological find uniquely and
within a spatial context. The geodatabase provides an extensible set of attributes, easily
updated to accommodate newly published GIS files and tables. This is a more uniform and
simple method to update data quickly and accurately than multiple separated GIS files and
tables, since options like domains to limit input to acceptable values or ranges exist. An
additional advantage to a geodatabase over a loose collection of separated tables, shapefiles

9

(.shp) or keyhole markup language (.kml) layers, or images is that use of a geodatabase can
facilitate SQL queries by providing a unified set of tables and establishing their relationships.
Finally, our solution can scale from a file geodatabase to an enterprise geodatabase if needed in
the future to allow for multiple users with versioning and access or editing controls.
1.5.1. Overview
This study will continue with the review of work of researchers from multiple disciplines
in chapter 2, to provide understanding for the types of analysis and data used to investigate the
key problems in human origins and migrations described in this chapter. The second chapter also
describes literature used to explore design and use of geodatabases in similar applications from
various related fields. Chapter 3 provides details of the methodology, including design and the
data sets used in the project. Chapter 4 shares the results of the geodatabase project, and ideas for
future work and desired improvements are outlined in chapter 5.

10

Chapter 2 Related Disciplines and Literature Review
The exploration of human dispersion and migration involves the disciplines of genetics,
biology, anthropology, archaeology, and history. Literature concerning the creation and
application of geodatabases in the service of research questions in these related fields was
essential. To better inform the process of building a geodatabase for studies of early human
populations, literature involving ancient peoples, their migrations, and genetic markers also was
reviewed.
Section 2.1 explores the questions and controversies that remain in the study of early
human origins and the use of genetics as one component of related studies. Sections 2.1.1 and
2.1.2 establish the use of Y-DNA and mtDNA markers as useful tools for current analysis of the
movements of ancient peoples and cement their inclusion in the data sets for this project. Section
2.2 outlines the use of geodatabases in a variety of scientific fields, providing models for this
project. Sections 2.2.1 and 2.2.2 specifically apply the use of databases to the question of
distributions in biology and to cataloguing archaeological samples, which have close parallels
with aspects of this work. Section 2.2.3 highlights the benefits of a digital approach in general,
and in particular one that enables the use of the Internet, to data sharing for historical research.
Section 3.1 describes the structure and beginnings of the Esri geodatabase format, and Sections
3.2 and 3.3 offer some insight into the benefit and practicality of selecting this format for this
project.
2.1. Genetics in Population Studies
Although the origins of our genus lie in Africa, the first development of the modern
form of humans, Homo sapiens, and its arrival on different continents continue to spark many
questions. Geneticists and anthropologists have looked at different types of evidence to find

11

answers to these questions and support various positions. For decades there has been some
support for a multi-regional hypothesis of humans arising from earlier related species, but others
opt for a somewhat similar “candelabra” hypothesis. Many others support the idea of a single
origin of modern humans differentiating themselves from earlier, related humans to become a
distinct new species, Homo sapiens (Hublin et al. 2017; C. B. Stringer and Andrews 1988; C.
Stringer and Andrews 2005; Wilson and Cann 1992).
Genetics work contributes strongly to the growth of the single origin theory (Gibbons
1997), as researchers can trace the origin of the so-called ‘Adam’ and ‘Eve’ of Y-DNA and
mitochondrial lines of all living humans and locate them in the African continent. However,
genetics studies reveal a combined ancestry of Homo sapiens and Homo neanderthalensis in
some regions (Pääbo 2014). The intermingling of closely related yet technically separate species
is part of the foundation of the multi-regional and candelabra hypotheses. Furthermore, work
involving agent-based modeling and migrations admits the possibility that human beings prior to
Homo sapiens could have crossed bodies of water to find food sources or settle (Baumann, n.d.;
Yang et al. 2010), not necessarily restricting the travels of earlier forms of our genus to slow,
overland treks. Although genetics has become an important part of investigation into early
human history, genetic analysis has not yet conclusively provided the answers to questions of the
first origins and expansions, and other disciplines and their methodologies will continue to play
an important role.
The single-origin or “out of Africa” theory usually ascribes an East African origin to
earliest Homo sapiens, then dispersion into other regions and continents, a journey as
summarized by Beyin (2011) as an origin in East Africa, then migration to Arabia, SE Asia, and

12

the Levant sometime after 150 kya. Eventually Europe and NW Asia would become inhabited.
Whether from one common origin, as Beyin (2011) describes, or from neighboring regions
already populated in part by our earlier ancestors and relatives, modern humans soon dispersed
across vastly different environments. Studies of these migrations frequently use genetic markers
in addition to other types of data to trace their paths out of regions and environments, such as the
use of hundreds of thousands of SNPS, rather than only mtDNA and Y-DNA haplogroups, to
reveal possibly three streams of Asian descent into the Americas at least 15 kya (Reich et al.
2012).
2.1.1. The Primacy of Nonrecombinant Markers
The uninterrupted “father line” and “mother line” genetic markers inherited, Y-DNA
and mtDNA respectively, are frequently the genetic markers used for tracing early human
journeys (Grugni et al. 2019). Although nuclear DNA is used to study the genetic evidence of the
early Homo sapiens, as in the Reich (2012) study of the Americas, and also is useful for tracking
hybrid population (e.g. offspring of Homo sapiens and Homo denisova) movements,
mitochondrial DNA (mtDNA) and Y chromosome DNA (Y-DNA) continue to figure
prominently in these kinds of geographic dispersal studies. These markers, particularly
examination of mtDNA haplogroup M dispersals, were used in Beyin’s (2010) examination of
Upper Pleistocene movements out of Africa. As Underhill and Kivisild (2007) remark, neither
mtDNA nor Y-DNA can be considered informative concerning the first speciation event of
modern humans from earlier types, but they remain “the most well resolved genetic loci for the
study of population histories since the out-of-Africa migration.” These types of markers have
been useful in following biological evidence of demic diffusion compared to the diffusion of

13

cultures, evidenced by characteristics of artifacts and languages, such as the question of Indo-
European (Gimbutas 1963).
2.1.2. Objections to Reliance on Non-Recombinant Markers
The ancient DNA of a group may be persistent, even if a direct father-to-son (Y-DNA)
or mother-to-daughter (mtDNA) lineage becomes extinct. These ancestors continue to pass down
other forms of DNA to descendants of the opposite sex for generations, allowing for earlier
ancestors to potentially be studied using nuclear DNA techniques (Reich et al. 2012).
Furthermore, Y-DNA and mtDNA account for only a small percentage of each human’s
individual genome. There are some limitations and valid concerns relating to exclusive reliance
on these types of genetic markers (Arias et al. 2018; Reich et al. 2012), but mtDNA and Y-DNA
remain important, established markers for analysis and will form the basis of genetic data for
inclusion in the geodatabase.
Regardless of the necessarily limited picture drawn from these lines, until techniques in
sequencing and analyzing ancient DNA improve these nonrecombinant markers likely will
remain prominent in research. Concerning analysis and genetic trees constructed of the other
types of nuclear DNA, Underhill and Kivisild (2007) point out that such trees may be robust
under some conditions. However, they also note that they are of low molecular resolution, which
provides a lesser understanding of our genetic history during “the pivotal past 100,000 years,
which is the time window of interest for most of the human migrations.” (Underhill and Kivisild
2007)

14

2.2. Geodatabases as Aids to Analysis
Many researchers have explored geodatabases within their respective areas to aid
analysis and visualization, in addition to providing a basic digital chronicle with a spatial
component. They have proven useful both in physical sciences, social sciences, and history. For
example, an Italian research team led by Viciani (2018) built a geodatabase for a botanical study,
and researchers Szabo et al. (2018) created a geodatabase for use in examining the history of a
forest. Archaeologists established numerous archaeological online databases - including some
spatial databases - for studies of ancient cultures in the Middle East. Rodriguez (2019) created a
geodatabase based on historical records to successfully explore trends in Mexican migration into
the United States. The study of ancient human migrations involves techniques from biological
sciences, archaeology, and social sciences, so successful use of a geodatabase in related fields of
study support its use in this project and provide models for design and implementation.
2.2.1. Geodatabases in Biology
Biodiversity or species distribution studies benefit from a geodatabase approach. The
idea of building a geodatabase to aid analysis of biodiversity in vegetation was proposed by
(Viciani et al. 2018) who expected to gain “…identification of spatial patterns of species richness
and of sampling effort...” among its benefits, as well as the examination of possible relationships
between species distribution and topographic factors. This approach was designed for the Parco
Nazionale delle Foreste Casentinesi, near Campigna, Italy, to support conservation studies, but
its methodology and research benefits could likewise be applied to studies of virtually any
species, either plant or animal, on land. For this vegetation study, (Viciani et al. 2018) the team
used roughly 680 reports from field studies using a hand-held GPS to collect spatial data on

15

vegetation species in the park in addition to literature data, much of which was lacking in
specific geographical data.
Viciani’s team obtained a digital elevation model (DEM) to investigate the topography
of the park alongside the various specimen data. Topography can be important to study of plant
distributions, since elevation, aspect, and slope can influence the growth of various species. It is
possible to imagine that the propagation of a plant species could be influenced by topological
conditions allowing or hindering seeds or roots to spread and increase the range of that variety.
The ability to incorporate DEMS or other raster files as part of analysis is useful not only in
botany, but other life sciences like anthropology, as demonstrated by Baumann’s work (2017) in
agent-based modeling of human migrations.
2.2.2. The “Internet Museum” and Digital Collections
The use of digital inventories with spatial data capabilities are not new to management
of archaeological finds. Many museums make use of EMu software by Axiell to keep inventories
of these artifacts and artworks. This allows institutions to share details of their collections with
others around the world and to maintain accountability for the locations of each item. IMu Maps
and IMu Object Locator allow for the mapping of objects in collections using floor plans (“EMu
– Collections Trust” n.d.).
Users can also use EMu for other diverse functions, such as tracking requirements for
repatriation of human remains samples, as required by some nations, or monitoring problems
areas that threaten the safe storage of objects in the collections (e.g. insect pests or mildew).
However, EMu users chiefly seem to use the system for collections management, and do not
appear to be using its developing spatial capabilities to carry out spatial analysis using the object

16

attributes filed in the database (Axiell 2018). This system, as currently implemented, cannot be
used in the place of a geodatabase for study of early genetics and migrations.
The use of geodatabases can aid researchers in locating considerable amounts of data
and easily visualizing it spatially, compared to other forms of web publication. Morrish and
Laefer as early as 2010 advocated for the use of web-enabled architectural heritage inventories.
Morrish and Laefer (2010) noted the considerable resources to compile inventories, resulting in
many “tomes” yet they were not fully utilized because of accessibility and ease of use.
In 1999, the historical preservation organization English Heritage began a project to
digitize and publish on the web. The system followed its original survey structure, with
independent entries lacking integration into a mapping interface or advanced inquiry tools
(Morrish and Laefer 2010). Monuments or sculptures commissioned during certain periods could
be specified with a search of the system, for example, but Morrish and Laefer (2010) pointed out
that its structure did not allow for a geographic visualization of their distribution. The
geodatabase format, like the one designed for this project, easily would allow for both these
queries and visualizations with proprietary and open source GIS tools.
2.2.3. Databases and Geodatabases in Archaeology
The approach of a database for archaeological research was investigated by Drzewiecki
and Arinat (2017). Drzewiecki and Arinat (2017) surveyed archaeological professionals in
Jordan regarding their use of databases for archaeology, and their concerns about utility, access,
and reliability. The survey by Drzewiecki and Arinat (2017) included all types of online
databases used in Jordanian archaeology studies, but at least one (MEGA-J) had full spatial
capabilities to display polygons. The use of databases had become very important to the

17

archaeologists, according to the researchers’ findings, notwithstanding some reservations
regarding reliability of the data. Most archaeologists would also conduct a literature review to
verify what was retrieved from the databases.
One important consideration illuminated by Drzewiecki and Arinat (2017) was that
improved locational accuracy could be a negative, in that the databases could serve as a
“catalogue of ‘looting-to-order’”. They added, however, that limiting access would only slow
down professional looters in their efforts to steal artifacts, not eliminate the problem. Drzewiecki
and Arinat (2017) summarized respondents in stating that the benefits of wide-spread and open
access to the databases outweighed the looting risk, an ever-present threat to sites. These
concerns account for the lower precision of coordinates used in this project, as journals typically
publish these because they would not increase this threat.
A site in Bethsaida, Israel, was the subject of efforts by Burrows in 2016 to create an
archaeology-specific geodatabase, designed to improve accuracy and utility when used by
personnel of various levels of skill or experience. Fieldwork at the site created an “unwieldy”
amount of data to record, according to Burrows, but modern digital methods were surprisingly
not deployed frequently to handle the problem. Burrows began constructing a geodatabase to log
the location of the artifacts discovered and accurately record characteristics in a systematic way
that could be easily reviewed using SQL queries. Both legacy data and new field study data went
into populating the database. The design and adoption of this geodatabase allowed use of
handheld GPS devices ((Burrows, n.d.) to be used to record more precise and accurate spatial
data and immediately describe basic attributes for the finds in the field.

18

This archaeogenetic geodatabase project necessarily will use less precise data, as it
currently relies upon the general locations of excavation sites in published papers. These seldom
offer the actual coordinates of each find, since as Drzewiecki and Arinat (2017) pointed out,
there is a real need to protect research locations, at least until work has been totally completed in
the field. However, the lack of precision for entries in this geodatabase is not an obstacle for the
project because unlike Burrows’ project (n.d.), it is not for the purpose of locating the sites or
additional specimens, but for spatially visualizing the patterns in related data on wider, regional
levels.
2.3. Developing the Geodatabase
2.3.1. Beginnings - Esri
The geodatabase (.gdb) is the native data format used by Esri, and is the usual data
format for editing and data management in Esri tools ArcMap, ArcGIS Pro, etc. Esri developed
the geodatabase (geographic database) to store both spatial and attribute data together
simultaneously in a single database management system. This integration of the relational and
object-oriented concepts in geodatabases allows its single management system storage for both
spatial data and attributes, according to (Twumasi 2002). Before this, the data model for
databases isolated geographic data from attributes of entities, for example ArcInfo used separate
files for attributes, called INFO tables, and spatial data was kept within indexed binary files
(Twumasi 2002).
2.3.2. Spatial Data Model
The geodatabase makes use of a spatial data model, which is divided first into either
raster/field-based or vector/object-based approaches. In turn, the raster approach may follow a

19

grid or regular tessellation scheme, or it may opt for an irregular scheme using variant sizes in
partitions. The vector approach may opt for an unstructured or “spaghetti” model or a structured
topologic model, the latter of which is more popular as it can offer a rich topology (Worboys
1995, Twumasi 2002) although it is less simple than an unstructured model.
2.3.3. Not Limited to Proprietary Format
The primacy of Esri in the GIS world is hard to deny, but many organizations have
begun to embrace other tools to support of open source, to have an alternative to expensive
licensing, and the ability to work in a more diverse set of formats, among other reasons. Some
believe that the geodatabase cannot be used with non-Esri tools, but this is not the case. Recent
versions of QGIS, past version 3.0, can read a geodatabase created by current versions of ArcGIS
(“Opening File Based Geodatabases in QGIS 2.4 • North River Geographic Systems Inc” 2014).
The Open Source Data Manager can read the directory using “OpenFileGDB” as type. As early
as 2015, GDAL versions were able to read and extract information from .gdb files (“Working
with File Geodatabases (.Gdb) Using QGIS and GDAL | Geospatial @ UCLA” n.d.), so QGIS
and other tools using GDAL can work with a geodatabase system. The ability to use the
geodatabase with other platforms makes it a suitable choice for a spatial database, since it can be
accessed by other popular applications if needed.

20

Chapter 3 Data and Research Methods
This chapter describes data and methods for design and construction of the geodatabase of
ancient peoples and genetic markers. This part of the project is divided into roughly four stages,
described in sections 3.1 through 3.4. The first stage described in 3.1 is the initial determination
of data needs and plans for data acquisition. Section 3.2 describes decisions made between the
various database options with spatial capabilities, and for software in general. Section 3.3
outlines the main design and concept of the database, including table relationships. The fourth
stage described in section 3.4 involves exploration of the data within the database structure and
testing the proof of concept. It outlines plans for SQL queries and basic visualizations to evaluate
its performance and fitness to its purpose. These stages initially were carried out in sequential
order but are repeated in a “feedback loop” as shortcomings or possible improvements to the
geodatabase were identified.
3.1. Data Needs and Acquisition
The data for the project consists of three main kinds: the archaeological find data (e.g. the
specimens themselves and their characteristics), male and female line genetic data, and
geographical data. The ideal was to have a wide scope and include global data for all three main
divisions. The human remains data were obtained from published papers by anthropologists or
archaeologists that included results of genetic analysis of mtDNA or Y-DNA haplogroups.
Reference tables for all the branches of the human genetic tree would need to be located or
created to include the genetic data in a normalized form suitable for an expandable database
project. Several public references from the International Society of Genetic Genealogists and
companies offering family DNA products (i.e. FamilyTree DNA) aided in the construction of
these genetic data sets when none could be located in a format suitable for use in a geodatabase.

21

The geography data was the easiest of the three to obtain, as multiple open source and Esri data
sets for countries, regions, and even past geological ages of the earth were available.
3.1.1. Spatial Context
A data set of world regions was included to provide a geographic context to the other data
sets. This ‘region’ data set came from Esri’s data portal. It is a shapefile of polygons with the
typical fields of OID, name, shape length, shape area, square kilometers, and square miles. Data
types include object ID, geometry, text, double, double, double, and double, in order. This will
be converted into a feature class for the geodatabase.
3.1.2. Finds – the Central Data Set of Ancient Human Specimens
The central data set, the finds, was anticipated to be the most difficult to obtain. Early
research showed that many papers described the needed data in narrative form, and did not
include tables, charts, JSONs, or formats that could be easily input into a database by automated
means. Furthermore, most papers described a single excavation site. These seldom contained
remains of more than five individuals, usually members of a family group. One exception was
the work by a team in 2017, Matheisen et al., who published a study on a whole collection of
hundreds of European finds and shared an Excel table with the designators, genetic data,
locational data, and many other important attributes to examine the temporal element, such as
identified culture or carbon dated estimates of age. Wherever possible, resources like this are
used to import data with less possibility of error than manual entry.
Since smaller and less easily formatted sources of data were the rule, work began
immediately on the data acquisition stage of the project. A timetable was set for 18 months to
locate and complete at least a basic compilation of the data, with a target of 300 entries in a form
for preliminary testing. Collection later exceeded this goal, with a total of over 2,700 entries for

22

the finds data set, more than 300 different Y-DNA clades (younger and usually smaller
haplogroups “downstream” from older, usually larger, “downstream” related haplogroups) and
over 340 distinct mtDNA clades or haplogroups. It is anticipated that some continued acquisition
and import into the database will be ongoing until the end stages of the project. A sample of the
initial spreadsheet used to compile data from published papers is provided in the table below.
The spreadsheet also used fields to record latitude and longitude coordinates for the find
locations (“findloc”) and notes on the culture period to which the find is attributed, but these are
omitted here for space considerations.
Table 1. Original data spreadsheet
Country findloc Desig Sex newY frmrnomY simpleY simplmt mtDNA
Austria Tyrol-Bolzano Oetzi M K1 K1 K1 G GL91
Bulgaria Vratitsa V2 M U U2e1'2'3
Bulgaria Krushare K8 M R R
Bulgaria Svilengrad P192-1 M U U3b
Bulgaria Stambolovo T2G2 M H H1c9a
Czech Republic Brandysek RISE568 F H H
Czech Republic Knezeves RISE566 M R-P310 R1b1a2a1aR H H2a
Czech Republic Brandysek RISE569 F H H1af2
Czech Republic Velke Prilepy RISE577 F T T2b

Many later additions to this central dataset of human remains were possible due to
discovery of a specialized OpenStreetMap. As of early 2019, OpenStreetMap (OSM)
contributors had created and published a global map of early human remains studied by
scientists. This map apparently incorporates several .geojson or .kml layers for different time
periods and regions, displayed in a map view with clickable pop-ups for related information, and
as such there is no mechanism for queries or filtered views. Each layer displayed presents not
only some facts about each find in the pop-ups but citations of the papers describing the related
scientific work. The inclusion of source citations in the OSM map allowed for quick location of
additional material for incorporation into the geodatabase. This was especially helpful to locate

23

data sets from areas where climate apparently had hindered early efforts at DNA extraction, and
so there was a much smaller body of published work with relevant data to locate. This resource
helped to expand the geodatabase to its global goal.
Details about the format of the map and files used in the OSM project could be seen
using the developer tools option in Google Chrome browser. Had there been many entries
included in the OSM project not already present in the geodatabase, it would have been possible
to acquire some of the .geojsons used in the OpenStreetMap project and to convert them into Esri
shapefile format, rather than performing a completely manual input from the cited papers.
Once the initial data phase concluded and the geodatabase was assembled, any similar
new finds could be incorporated. For any ongoing data collection, new material could be merged
with the existing ‘archaeological_find’ feature class once it was converted into new shapefiles
from tables or other common GIS files types. Using ArcGIS Pro, it is possible for field data to be
amended or added to fit the schema of the geodatabase and updated with the data management
tools.
The initial data acquisition phase resulted in a point shapefile called
‘archaeological_find’, now converted into a feature class for inclusion in the geodatabase. Figure
1 given below shows the table view in Arc Catalog immediately after its initial conversion into a
feature class, prior to eliminating unneeded fields and renaming others for consistency in the
database design. More details about this feature class and others will be provided in section 3.3
concerning the methods of the geodatabase.

24

Figure 1. Table view in ArcCatalog after feature class conversion

3.1.3. Genetic Data – the Family Lines of mtDNA and Y-DNA
The genetic data in the finds database was used to create the two data tables for the
mtDNA and Y-DNA genetic haplogroup. Unique instances of the ‘archaeological_find’
subclade, the “mtDNA” or “Ydna” fields shown above, were found using Excel to create the
basis of the two tables. The most common clades also were added. The resulting columns were
copied and used as the first column field of two new tables. Although the tables will expand to
include many clades not yet present in the finds dataset, this step ensures the presence of the
most relevant groups for analysis and queries in the finds for testing from the beginning.
After the initial list of Y-DNA subclades or clades was made, an additional field was
created for the SNP nomenclature of those groups, if determined. The haplogroup to which the
clades belong is another field, and the haplogroup’s direct ancestor is the last field included in
the table. The mtDNA table follows an identical structure, except that a separate SNP
nomenclature is not used for mtDNA groups. Both tables were created as .csv in Excel, then

25

imported into the geodatabase. Except for the object ID, all fields are ‘text’ data types for this
schema. The details of these fields in the design will be discussed further in section 3.3.
The consistency in mtDNA naming conventions makes working with the data much
simpler, especially when designing the tables for queries. Accommodating a changing and
inconsistent system of nomenclature for Y-DNA data has posed several challenges to the correct
import of the data as well as construction of the tables and relationships. The solution at this
point is to use the International Society of Genetic Genealogists (ISOGG) naming conventions
for most purposes, but to keep an SNP field in the table for cross-referencing. They may possibly
be fully incorporated later when these SNP designators are universally accepted and
standardized.
3.2. Software and Tools
Software offers a variety of potential solutions for this spatial database project. Several
requirements are present for a geodatabase of archaeogenetic data from around the world. A
chief concern is the ability of the software to support complex analysis and to offer visualizations
that can differentiate between many objects or characteristics. The data for the specimens
consists of numerous material cultures identified by researchers, a wide range of dates, and
hundreds of clades or subclades, among many other attributes included for visualization and
analysis. The data for the specimens alone could grow to number several thousand entries, but
storage needs are still minimal compared to many commercial database applications.
Nevertheless, the database must handle potentially complex queries well. It is also extremely
important to be able to update and edit the data easily, since new techniques are making more
samples available with DNA analysis and the field of genetics is changing rapidly.

26

3.2.1. Prototype and Final Software Solution
The initial trial of this database was created in Microsoft SQL Server Management Studio
and was a fully functional prototype. A test was also done of the database using Access. SQL
Server Management Studio makes database management relatively simple and can accommodate
spatial data. However, the capabilities of visualization and analysis offered by other tools are not
available in SQL Server Management Studio, so it was not considered a viable option for this
project. Other options, including several open source software solutions, were explored until the
choice was made to construct this spatial database in a geodatabase (.gdb) format using Esri.
3.2.2. Software Solutions for Complete Spatial Database
ArcGIS Pro by Esri was selected as the software of choice after reviewing some
comparisons of other database management software with spatial capabilities, such as the
comparison of MySQL and PostGIS capabilities by Piórkowski (2011), as well as considering
personal experiences with the performance of the various tools available for database
management (DBM) and GIS. A key factor in this decision is the simplicity of a unified solution
for all DBM, GIS, data sharing, and cartographic needs. As Rodriguez (2019) points outs in his
work creating a geodatabase for historical migration data, using a comprehensive software
program that can store, handle, and manage both spatial and nonspatial data is essential.
ESRI’s GIS software and geodatabase format not only provide these requirements, but
also offer the benefit of a widely used platform with rich visual displays, user-friendly
cartographic layouts, and many advanced analytic tools at the user’s disposal. This toolkit uses a
largely intuitive GUI that allows for quick setting of parameters and can reduce typos or coding
language errors. Esri has the additional advantage of providing a data portal to make ongoing
data acquisition and sharing between researchers easier.

27

Besides Microsoft Excel for compiling notes and initial data preparation, ArcGIS Pro
version 2.3 is the main tool for this project. Tables and shapefiles can be imported into the
geodatabase and converted into feature classes or geodatabase tables. Relationship classes may
also be created with ArcGIS Pro. If domains or field changes are needed during the project, it
can be used to make them. The multifunctionality and highly satisfactory performance of both
tools have fully met the software needs of the project.
3.3. Database Design
To move beyond a simple collection of shapefiles and into a geodatabase, it was
necessary first to construct a relationship diagram. The design took a few different shapes before
the ERD was finally constructed after some trials and feedback. Following the design with
normalized tables, some initial fields from ‘archaeological_find’ were removed after it had been
converted into a feature class in the geodatabase. The initial ERD is provided in figure 2 below.
It is anticipated that additional adjustments will be made after the database is constructed and
further tested.

28

Figure 2. Conceptual ERD

3.3.1. Overview of Tables
The current design includes four parts, two feature classes containing spatial data types,
‘region’ and ‘archaeological_find’, and two genetic data tables, ‘yDNA’ and ‘mtDNA’, that
contain only non-spatial data. Ideally, all table names would use either singular or plural forms
and letter case to avoid needless confusion during queries or any scripting involving the tables.
Field names also were somewhat constrained by Esri requirements, such as avoiding spaces. To
aid clarity, the names for DNA-related tables retained the upper-case form, and these were the
only exceptions to the lower case and singular forms used in the geodatabase. Again, for clarity

29

the name of the central feature class of samples from human remains is called
‘archaeological_find’ to provide more immediate understanding of the contents.
In the geodatabase design, the central table is the feature class for the archaeological
finds themselves. It contains entries for the human specimens and their attributes, including the
mtDNA and Y-DNA clade or subclade identified, and spatial data expressed as geometry. For
this data set, the geometry is point data derived from medium precision and accuracy latitude and
longitude coordinates, which was originally converted using XY Table to Point tool. Two other
tables contain mtDNA data and Y-DNA, respectively, and are named ‘mtDNA’ and ‘yDNA’.
These tables both contain fields for names of each clade or subclade, simply called either
“mtDNA” or “yDNA”, the upstream “haplogroup” to which it belongs, and the haplogroup’s
direct ancestral haplogroup, called “ancestor”. The ‘yDNA’ contains an additional ‘alty’ attribute
of alternative nomenclature. The ‘region’ table has the “region” or name, its polygon
“geometry”, and the standard Esri fields for “shape_length”, “shape_area”, “square_kilometers”,
and “square_miles”.
3.3.2. Relationships and Cardinalities
The relationships for these tables are relatively simple and can be assigned using ArcGIS
Pro. Each entry in the ‘archaeological_find’ table must be linked by its ‘mtDNA’ field to exactly
one entry in the ‘mtDNA’ field of the ‘mtDNA’ table, since specimens without DNA analysis
were not included in the project. However, each entry in the ‘mtDNA’ field of the ‘mtDNA’
table may have several ‘archaeological_find’ entries that belong to it. Therefore, the cardinality
of the relationship of ‘mtDNA’ to ‘archaeological_find’ is “one-to-many”, for each
‘archaeological_find’ specimen belongs to exactly one ‘mtDNA’ entry but a single ‘mtDNA’
entry could have many specimens that belong to it. Likewise, entries in ‘archaeological_find’

30

may only have one single corresponding ‘yDNA’ value, but again each ‘yDNA’ table entry may
also have many entries in the ‘archaeological_find’ table that belong to it. The cardinality
therefore is “one to many” from the ‘yDNA’ table to the ‘archaeological_find’ table.
3.3.3. New vs. Original Design for Tables
The early prototype structure had been overly complex, so it was simplified in the new
design for Esri’s geodatabase format. Keeping only a ‘region’ data set and not including a
‘country’ data set, which would not be relevant for the temporal span of the geodatabase,
reduced the number of tables. There is no effort to connect DNA groups to regions where they
are commonly found today. This is because the aim of the geodatabase is to allow for analysis of
these relationships as they existed in the past, and so it was beyond the scope of this project.
Instead, a ‘region’ feature class serves as an example of using geographic references or other
spatial data in queries on the central feature class ‘archaeological_find’ and the DNA tables. The
relationship of ‘region’ to ‘archaeological_find’ will utilize spatial operations rather than explicit
keys in a departure from relational database design in the geodatabase design.
3.3.4. Data Types
Operations may fail if the proper data types and lengths are not specified. For example,
dates in some data sets may be formatted as text, or a field that may require a decimal to the
hundreds place for some records may be formatted as an integer; these would require changes to
the data type to be used as intended. For this geodatabase, the estimated earliest and latest ages
of the archaeological finds were included to provide further context and allow for easier queries
involving multiple cultural periods or groups. However, they were left as integers, because of
difficulties in working with such early and imprecise dates. Positive and negative integers were

31

used for CE dates and BCE dates, respectively, to not only facilitate queries on the finds from
different eras but to allow the user to perform age calculations.
Aside from object IDs and the integers for estimated ages, most other data types in the
schema are text, or varchar (255), except for the two feature classes that contain geometry. The
‘region’ entity additionally retains the four length and area related fields from Esri, which are
double. The details of data types in the schema of the current geodatabase plan are illustrated by
figure 3.

Figure 3. Schema drawing with data types of early geodatabase plan

3.3.5. Region Feature Class
The ‘region’ feature class was made from a data set created and provided by Esri,
consisting of polygons representing common regional divisions of the world’s land masses. A
relationship class using key fields is not strictly necessary with ArcGIS Pro or other GIS because
of spatial join capabilities. One region may be the location of many archaeological finds, and

32

each find will only belong to one region, but the cardinality is not shown here by the crow’s foot
notation for figure 3 because the tables ‘region’ and ‘archaeological_find’ were not formally
linked. Although this data set was included as a permanent feature class in the geodatabase, keys
were not used to create a relationship class in order to demonstrate this capability. The use case
scenario includes analysis of the archaeological finds alongside historical geographic data, e.g.
ancient coastlines or glacial extents. Many such data sets are available but are localized, and so
were not incorporated into the geodatabase at this time. The ‘region’ feature class will show how
this analysis of finds with respect to geographic areas can be accomplished in a GIS with a
geodatabase, using spatial join, without the need for creating special key fields as would be
required in a relational database.
3.3.6. Archaeological_find Feature Class
‘Archaeological_find’ is the central feature class of the geodatabase. As described in the
data section 3.1, the ‘archaeological_find’ feature class was created using Arc Catalog from a
point shapefile made in ArcGIS Pro. Its attribute table includes the geometry, along with other
non-spatial attributes including the designator, “culture” indicating the material culture or period
(e.g. Neolithic, Hittite, Magdalenian, etc.), “earliest_age” giving the specimen’s earliest
estimated age, “latest_age” giving its latest estimated age, mtDNA clade or “mtDNA”, the Y-
DNA clade “yDNA” in ISOGG form if specimen is male and this data could be obtained,
alternative nomenclature “SNP” for the Y-DNA clade if available, and “citation” for a doi –
useful to quickly verify the data or obtain more information concerning the specimen in the
original paper. Each entry in ‘archaeological_find’ belongs to exactly one ‘mtDNA’ entry as
shown in the crow’s foot in figure 3, but belongs to one or zero in ‘yDNA’ since many samples
are genetically female. The ‘mtDNA’ and ‘yDNA’ entries may both have many

33

‘arcaheological_find’ entries that belong to them. Details of the fields and data types as shown in
ArcGIS Pro’s Catalog can be seen in Table 2 outlining the ‘archaeological_find’ feature class
alone. Its geography view provided immediately following in figure 4.

Table 2. Data types for ‘archaeological_find’

Field Data type
ObjectID object ID
Shape geometry
earliest_age long
latest_age long
culture text
designator text
mtDNA text
yDNA text
citation text

Figure 4. Geography view of feature class ‘archaeological_find’ points in Catalog

34

3.3.7. Genetic Data Tables
As introduced in section 3.1 on data needs and acquisition, the tables for mtDNA and Y-
DNA contain fields for the DNA clades, their haplogroups, and the direct ancestor of the
haplogroups. These progress from left to right from the younger, downstream subclade or clade
to the haplogroup to which it belongs. In turn, the haplogroups’ direct ancestor, or further
upstream haplogroup, is listed. This additional field is included to assist in querying for related
groups and keep the database design as simple as possible. No regional associations for the
subclades or haplogroups of either mtDNA or Y-DNA are assumed, although there are frequent
references to them in the literature. The primary keys for the ‘mtDNA’ and ‘yDNA’ tables
remain the “mtDNA” and “yDNA” fields to connect the ‘archaeological_find’ table through the
“mtDNA” and “yDNA” fields as foreign keys. If junction tables do become necessary for any
future changes to the geodatabase and its uses, the Object ID fields could be used to set up
primary and foreign keys with simpler numerical IDs rather than text. Data types for the two
genetic data tables are all text, except for the Object IDs. An example from mtDNA showing
several entries is given below in table 3.

35

Table 3. Sample of mtDNA genetic clades and upstream haplogroups

OBJECTID haplogroup mtDNA ancestor
1 A A N
2 A A1 N
3 A A11 N
4 A A14 N
5 A A15 N
6 A A16 N
7 A A17 N
8 A A1a N
9 A A2 N
10 A A2a N
11 A A2b N
12 A A2c N
13 A A2d N
14 A A2h N
15 A A2i N
16 A A2p N
17 A A4f N

3.4. Testing the Adequacy of the Data and Functionality of the Database
The fourth phase of the methods process will evaluate the data and test the database’s
performance. The tests of the data will determine if it is possible to complete some basic
analyses with the sample size and the attributes included. These queries will test the relationship
classes set up in the geodatabase, ability to select multiple attributes and limit to only the desired
results, and the spatial join in place of a standard relationship between geographic data sets and
tables or point data feature classes. These will also illustrate some benefits of working within a
GIS that can render maps with flexible options. The details of these short tests are as follows:
• A spatial join will be run on the ‘region’ and ‘archaeological_find’ data sets, then
a SQL query will locate the regions where Q haplogroup-descended specimens
are found in the data.

36

• The second example will test the relationship of ‘mtDNA’ and
‘archaeological_find’ and explore the ability to show the temporal-spatial
distribution of an early mtDNA haplogroup. First, using a query the database will
find all members of clades or sub-clades belonging to haplogroup M. Then using
a classed color ramp in ArcGIS Pro’s symbology options, the various age ranges
of those related ‘archaeological_find’ samples will be mapped in ArcGIS Pro
across the ‘region’ data with an underlying basemap for context.
• The geodatabase will be queried again for the ‘yDNA’ relationship and for
multiple attributes. The example query will find results restricted using the
“earliest_date” field for two branches of a larger haplogroup. For this test, the
“yDNA” belonging to R1a and R1b will be located in the ‘yDNA’ table, and
related points from archaeological_find will be queried in turn to find those older
than 1 CE. The results will be mapped and visualized to show the areas with
concentrations of the two branches.
In addition, several other SQL queries to check the geodatabase’s functionality and
success of the basic design will be run. The results will be presented in simple tables.

37

Chapter 4 Results
This chapter will describe the final design of the geodatabase and the reasoning behind major
changes or modifications in section 4.1. A benefit of the geodatabase for spatial data over
standard relational database structure is briefly described in section 4.2. This chapter will show
the results of queries that demonstrate the functionality of the geodatabase, as implemented using
Esri ArcGIS Pro, in section 4.3.
4.1. Final Database Design
The final physical structure of the geodatabase changed only slightly from the original
concept, due to some differences in Esri’s file structure compared to requirements of the standard
normalized database structure. Some fields were removed from ‘region’ feature class obtained
from Esri and one added into the ‘archaeological_find’ table. Another difference from the
original design was that domains for better data integrity were introduced for the ‘yDNA’,
‘mtDNA’, and ‘archaeological_find’ tables. The use of spatial capabilities to avoid explicit keys
linking the ‘archaeological_find’ to ‘region’ feature classes was successful and maintained.
4.1.1. ERD Changes
Overall, the differences in the structure of the conceptual ERD described in Methods
Chapter 3 versus the end result were minor. The final form of the ER diagram is shown below in
Figure 5. Due to the changing nomenclature for Y-DNA, a “yDNA_notes” field was created to
document regarding data for ‘archaeological_find’ in case expansions or changes were needed in
future. However, unneeded fields from the ‘region’ feature class included by Esri that could be
derived from its “geometry” were eliminated, namely “shape_length”, “shape_area”,
“square_miles” and “square_kilometers”.

38

Figure 5. Conceptual ERD as implemented

4.2. Moving Away from a Standard Relational Database
Using a GIS for spatial database creation and management as well as analysis allows us
to make use of processes such as spatial joins. The spatial join locates entries that fall within
specified regions or boundaries, creating a layer with these relationships. This layer then can be
further queried. Any other geographic tables potentially added in future to the geodatabase would
not require a relationship set up using keys, as is necessary in a standard relational database if the
database creator takes advantage of this capability. The results can be saved and added as a table
or feature class within the geodatabase, but the spatial join itself does not remain a permanent

39

part of it. For this reason, the ‘region’ table represented in the drawing of the ERD provided
above in Figure 5 reflects the relationship of ‘archaeological_find’ to ‘region’. The inclusion of
the ‘region’ table within the database, but without the need for directly relating the tables using a
key field, is described further in 4.2.1 and illustrated in a physical diagram of the final
geodatabase form in Figure 6.
4.2.1. Final Schema as Implemented in ArcGIS Pro
The final physical form of the tables in the geodatabase was consistent with the earlier
plan discussed in Methods but eliminated the extraneous fields for shape length, area, and
measurements of square kilometers and miles. These were not used in the geodatabase and since
they could be derived from the geometry of the feature class, they were unnecessary and
redundant. Only “region” and “geometry” fields were retained in the final implementation of the
design for ‘region’. As planned, for ‘archaeological_find’ and the DNA tables, Catalog
established relationship classes with cardinalities and keys, and spatial operations were used
instead to link ‘region’ and ‘archaeological_find’. Data types for all the remaining fields in all
tables were maintained as in the schema described previously in 3.3.4. Details of these data types
and the cardinalities of relationships are illustrated in the crow’s foot notation diagram for the
final version of the physical database given in Figure 6 below.

40

Figure 6. Schema drawing with data types for geodatabase

4.2.2. Aspects of Relational Database Structure
ArcGIS Pro was able to establish the relationships shown in Figure 7. Although the
spatial tools made some aspects of relational database unnecessary, previously discussed in
4.1.1., Esri’s geodatabase format should still have a structure to allow successful SQL retrievals.
The “Clause” mode for queries presents a very intuitive interface for users unfamiliar with
formal query language through guided selections of parameters. However, an example of the
SQL involved in the actual operation can be seen by switching away from the default “Clause” to
the “SQL” mode in a Definition Query. When performing a query on entries in
‘archaeological_find’ it is possible to retrieve not only the DNA clades, either ‘mtDNA’ or
‘yDNA’, but also the related haplogroups and their direct ancestors from the two DNA tables. A
geodatabase provides the structure to use an established relationship classes to facilitate queries
on normalized tables. Examples of this functionality are shown later in section 4.3.

41

In ArcMap and ArcGIS Pro, “Create Relationship Class” is one of Esri’s Data
Management Tools. The tool asks for the origin and destination tables, the origin primary and
foreign keys, whether the relationship is simple or composite, and the cardinality. After it runs, a
relationship class will appear in the geodatabase itself. An example immediately follows in
Figure 7.

Figure 7. Create Relationship Tool and parameters

The successful creation of a relationship and its properties can be verified in Catalog as shown in
the following two figures, Figure 8, and Figure 9.

42

Figure 8. Relationship Classes as viewed in Catalog

Figure 9. Relationship Class properties as seen in Catalog

43

4.3. Resulting Queries and Visualizations
4.3.1. Query with Spatial Join Test – DNA Tables and Region
The first result in testing the functionality of the geodatabase is the use of the spatial join
to query without a formal relationship of a geographic table to other tables in the database. The
example query uses a spatial join of the ‘archaeological_find’ feature class to the geographic
reference feature class ‘region’. The results of this spatial join were then queried in ArcGIS Pro
using “Definition Query” to locate the regions where subclades belonged to the Q haplogroup.
In the early ages included in the geodatabase, Q-descended subclades were found in East,
Central, and West Asia, all regions of Russia, East Europe, and both North and South America.
Today, descendants of pre-Colombian populations in the Americas almost entirely – as much as
90% - belong to the larger Q haplogroup (Grugni et al. 2019) with the remainder of the modern
distribution of the haplogroup principally in northeast Asia. This example of a spatial join query
displays the ancient distribution of Q haplogroup Y-DNA entries in nine regions, Northern
America, Central America, South America, Eastern Europe, European Russia, Central Asia,
Eastern Asia, and Southern Asia. This is shown by the solid-colored regions below in Figure 10.

44

Figure 10. Spatial join query, Q Y-DNA haplogroup

4.3.2. Relationship Class Test – Mapping Related Data
The second test of the geodatabase verifies that the relationship class structure allows
retrieval of related data from a table for queries on the ‘archaeological_find’. The relationship
class between ‘mtDNA’ table to ‘archaeological_find’ table is used to locate descendants of the
mtDNA haplogroup M, the matrilineal or “motherline” marker, from ‘archaeological_find’ and
display their distribution. The tables were normalized, so ‘archaeological_find’ contains only the
mtDNA clade codes in the field “mtDNA”, which serves as a key to the other data in the
‘mtDNA’ table. The ‘mtDNA’ table first was restricted by a definition query to the M
haplogroup, then related records in the ‘archaeological_find’ feature class were selected. After

45

this step, the multiple entries in ‘archaeological_find’ highlighted only belonged to clades
belonging to haplogroup. Furthermore, the query selected all clades for the M haplogroup,
demonstrating that using the relationship class between ‘mtDNA’ table to ‘archaeological_find’
was successful. For clearer visualization, those related records were saved as a separate layer,
shared below in Figure 11.

Figure 11. Query of finds and related mtDNA table, M mtDNA haplogroup

46

4.3.3. Test of Querying against Multiple Fields – Y-DNA Examples and Date Fields
The geodatabase allows for queries of multiple fields to find results that meet a limited
set of criteria. The geodatabase was queried to find only members of clades of the R1a and R1b
Y-DNA haplogroups, the patrilineal or “fatherline” markers, for a map showing where the
respective haplogroup branches were concentrated. Because R1a and R1b distributions today
differ in frequency between eastern and western parts of Eurasia (Underhill et al. 2015), the
example also checks whether the geodatabase structure can accommodate queries to investigate
and compare distributions of the haplogroups in earlier millennia.
For each haplogroup, the ‘yDNA’ table first was queried to find either R1a or R1b.
Second, the related records in ‘archaeological finds’ were selected. The query was then further
restricted to only return ‘archaeological_find’ entries dated prior to 1 CE, using where
‘earliest_age’ is less than or equal to 1 CE, or “earliest_age <= 1” as displayed in Esri’s SQL
box. This example also served as a test of the relationship class on the ‘yDNA’ table to
‘archaeological_find’. Each query result was exported as a separate file.
The R1a and R1b results are shown in Figure 12 following this paragraph. R1a and R1b
samples are visualized with blue and pink, respectively, with darker areas appearing where one
of the two branches were more concentrated, and purplish areas appearing where there likely
were concentrations of both branches. These maps illustrate that R1a and R1b did show different
distributions from one another in the past, similar to what is seen today.

47

Figure 12. R1a and R1b distributions before 1 CE

R1a in ancient times had a stronger presence in eastern Europe and central Asia, to the
northeast of the Altay mountain range and south to parts of present-day Mongolia, than R1b. R1a
also appears to possibly have been notable in the Baltic areas and parts of Scandinavia prior to 1
CE. Today, R1a groups mostly appear in the eastern regions of Europe and still are noted in
Central and South Asia (Underhill et al. 2015).
The most westerly areas of Europe today have a high percentage of R1b-related Y-DNA
groups (Underhill et al). R1b query results show that in ancient times R1b was already

48

predominant over R1a in westernmost areas of Europe. Although R1 groups are not the
predominant ones seen in Scandinavia today, but those commonly noted in the region now are
usually of the R1b branch versus R1a. It appears this may not have been the case in ancient
times, as the Baltic and parts of Scandinavia show many samples for R1a but not R1b. In ancient
times R1b also was seen, alongside R1a, in the steppe areas of eastern Europe, and as far east as
the environs of the present-day city of Urumqi.
4.3.4. Make Query Table – an Alternative to Standard Joins and Definition Queries
A more general query tested locating the oldest finds by using “Make Query Table” tool.
This is an alternative to creating joins or definition queries in the map view that can be used
straight from the analysis toolbox. A map layer can be created rather than solely a data table of
results if one of the tables includes a geometry field, or “Shape” as named in Esri. Selecting
fields is not necessary in the tool except when using Model Builder; the selections below serve as
illustration only. A screenshot of the “Make Query Table” interface follows immediately in
figure 14. This example shows that by using SQL “archaeological_find.earliest_age <= -8000”,
the GIS could retrieve all the finds at least 10,000 years old. A sample of the results table from
the query is provided in table 4, showing some of the oldest archaeological_finds and attributes,
immediately following figure 13. Table 4 has omitted OID and geometry fields for space
considerations; ‘yDNA’ was returned successfully in the query but the top ten oldest selection of
results shared here are all from females.

49

Figure 13. Make Query Table Processing Parameters interface

Table 4. Sample of Make Query Table results for samples older than 10 kya

earliest_age designator culture mtdna citation
-38000 Tianyuan Tianyuan B http://www.pnas.org/content/110/6/222
-29250 Krems WA3 Gravettian U5 http://www.nature.com/nature/journal/v
-25800 Ostuni1 Gravettian M http://www.nature.com/nature/journal/v
-16800 El Miron Magdalenian U5b http://www.nature.com/nature/journal/v
-15000 AfontovaGora3 AfontovaGora3 R1b http://www.nature.com/nature/journal/v
-13000 TAF012 Iberomaurusian U6a http://science.sciencemag.org/content/e
-12250 Oriente C Late Epigravettian U2' https://www.sciencedirect.com/science/
-10000 CB13 CB13 K1a http://mbe.oxfordjournals.org/content/e
-9650 USR1 Denali complex C1b https://www.nature.com/articles/nature

Several other queries were carried out using the “Make Query Table” to check the
structure of the ‘yDNA’, ‘mtDNA’, and ‘archaeological_find’ tables. This tool does not appear
to allow for adjustments in the SQL mode to switch from inner and outer or left and right

50

queries. Instead, tables should be selected in the appropriate order when filling out the
parameters, joining can be accomplished by matching the corresponding fields. For example, the
SQL
archaeological_find.yDNA = yDNA.yDNA And (yDNA.ancestor = 'BT' Or
yDNA.ancestor = 'A00' Or yDNA.ancestor = 'B' Or yDNA.ancestor = 'A1' OR
yDNA.ancestor = 'A1b' OR yDNA.ancestor = 'CT')
yielded results for the male finds from all cultures and eras descended from these earliest Y-
DNA haplogroups in the geodatabase. A check of the ‘archaeological_find’ table “Y-DNA” field
confirmed that these results included the clades or other Y-DNA groups that descended from
these groups present in the geodatabase. The table created by the query, with “citation” field
omitted for space, is given here in table 5.
Table 5. Make Query Table results from earliest Y-DNA haplogroups

OBJECTIDculture designator yDNA ancestor haplogroalty mtDNA earliest_alatest_age
1511 Hunyadihalom I2783 CT BT BT T2b -4228 -3963
2043 Straubing(Early Bronze Age WEHR_1474 CT BT BT V -2029 -1772
2044 Straubing(Early Bronze Age WEHR_1564 CT BT BT V -2029 -1772
1279 Pre-Pottery Neolithic ZMOJ=BON014 C CT CT K1a -8300 -7800
1357 Neolithic I0706 C CT CT K1a -6300 -5900
1376 Hoabinhian La368 C CT CT M5 -6000 -5850
1379 Starcevo I3498 C CT CT U8b -5837 -5659
1407 Early Neolithic Kitoi DA357 C CT CT A+1 -5500 -5000
1466 Lengyel I1899 C CT CT T2b -4800 -4500
2631 Medieval R1285 C CT CT T2c 771 1490
2680 Medieval Nomad DA106 C CT CT C4b 1000 1250
2702 Kipchak DA23 C CT CT F1b 1045 1095
1289 Nachikufu(?) I2966 BT A1b A1b L0k -8000 -3000
2041 Straubing(Early Bronze Age WEHR_1414 BT A1b A1b K1a -2029 -1772
2247 Philistine ASH008.A0101 BT A1b A1b H2c -1000 -900
2215 late Stone to Metal Age 4/A B2b B B L1c -1260 -1020
1365 early Stone to Metal Age 2/SE I B BT BT L0a -6020 -5740
2342 Pastoral Neolithic I8804 A1b1b2 A1 A1 A-L427 L4b -751 -411
2416 Ballito Bay B Ballito Bay B A1b1b2 A1 A1 A-L427 L0d -199 18
2433 Ballito Bay Ballito Bay A A1b1b2 A1 A1 A-L427 L0d -36 119
2341 Pastoral Neolithic I8758 A1b1b2a A1 A1 A1b(xA1b L0a -751 -415
2429 Later Stone Age I9133 A1b1b2a A1 A1 A1b(xA1b L0d -50 200
2505 Hunter-gatherer I9028 A1b1b2a A1 A1 A1b(xA1b L0d 300 0
2736 Pastoral Neolithic I89194 A1b1b2b A1 A1 A1b1b2b; L4a -8000 10000
2676 Medieval urm035 BCDEF BT BCDEF H2a 900 1200

51

Another test query involved finding all finds prior to 1 CE belonging to the V
haplogroup:
archaeological_find.mtDNA = mtDNA.mtDNA And mtDNA.haplogroup = 'V' And
archaeological_find.latest_age <= 0
This SQL returned the pre-Common Era finds in ‘archaeological_find’ with all the clades
belonging to the upstream haplogroup V, shown immediately below in Table 6.
Table 6. Make Query Table restricted date results for mtDNA haplogroup V
designator culture mtDNA yDNA haplogroup ancestor
I2012 Roessen V1a V HV
N27 Lengyel(BKG) V14 V HV
N19 Funnel Beaker (TRB V14 V HV
I10564 Afanasievo V1a V HV
I7290 Bell Beaker V3 V HV
I5367 Bell Beaker V10 V HV
I2365 Bell Beaker V3 R1b1a1a2a1a2b1 V HV
I7638 Early Bronze Age V10 I2a2a V HV
OTTM_156 Middle Bronze Age V1b K2b2a2 V HV

The last two queries for V haplogroup and early Y-DNA ancestors A00, A1, A1b, B, BT,
and CT also yielded results with geometry that could be seen in the map view. This part of the
tool worked well as described in Esri literature but did not display the types of joins it was
performing. As stated earlier, there does not appear to be a way to use the SQL window to
change to OUTER RIGHT JOIN or make similar specifications on the kind of join. This may be
preferred by many users, but others may be disappointed by the limit. (The tool was not tested
using the Python scripting environment; it is possible that different kinds of joins are allowed
there.)
Esri also indicates that the table created by the tool should be exported to retain it
permanently, but this generated an empty table on two test queries. In these instances, a

52

workaround was achieved by selecting the original table results. Then, one may create a layer
from the selection, and finally using data output save either a shapefile or table.
The query results that could be exported successfully did not seem to differ in any
meaningful way from the ones that failed. It is unclear why this happened, but ArcGIS Pro
occasionally failed to initially load the entire set of rows, and the attribute table would have to be
reopened. It is likely that use of a OneDrive directory to output analysis results had a role in the
inconsistent behavior. Working locally when possible, then uploading results to cloud-based
storage for back ups or sharing is advised.

53

Chapter 5 Conclusions and Discussion
This chapter discusses the results of the geodatabase project, lessons learned, and ideas for future
related work. Section 5.1 provides an overview of the successes and shortcomings of the
geodatabase project. Reasons for insufficient data in some areas of the world are discussed in
depth in 5.2, along with the potential for these issues to be addressed. Parts of the geodatabase
project design and process that worked well or potentially could be improved will be addressed
as a “lessons learned” in section 5.3, including the use of domains in some fields to limit error.
These possibilities for future changes, especially concerning open source alternatives to confront
budget and/or license constraints, will be discussed in the final section 5.4.
5.1. Overview
The design and implementation of the geodatabase in Esri’s ArcGIS Pro was a success,
and multiple queries and visualizations could be easily made with the geodatabase. The
relationship classes made it possible to find related information and used normalized tables in the
GIS. The initial creation of tables and feature classes in the geodatabase went smoothly.
Domains with coded values were not difficult to implement. ArcGIS Pro did sometimes crash in
very simple processes of editing when using domains but did not usually provide an error to aid
troubleshooting. However, no work was lost, and the domains and other aspects of geodatabase
structure were intact when the GIS was restarted.
A key issue that arose early limited the utility of the geodatabase in global analysis of
some of the key questions in studying early human evolution and migrations; namely, unequal
sample sizes for different continents. This was one of two external issues that presented
challenges for this project and were distinct from design and technical questions. The other issue

54

was the evolving system of nomenclature for Y-DNA clades and haplogroups, and a working
solution was quickly found. These external issues are expected to partly resolve themselves and
become far less of a challenge with time, although there are multiple reasons for unequal
sampling and the issue is complex.
The growth of the field of genetic genealogy has created a happy dilemma in which
progress is happening so quickly that a consensus has not been achieved in some details such as
nomenclature or naming conventions. The mtDNA naming conventions appear to be more stable,
but in Y-DNA at least three methods have been noted to name the mutations that mark a
departure of a clade or subclade from its parent branch. Two are commonly in use, and it remains
to be seen which nomenclature eventually may come to predominate completely.
Until that time, the geodatabase makes use of the ISOGG more commonly seen in the
papers used in data collection. The problem of Y-DNA nomenclature being in flux was
overcome by using this method as the primary and using resources to “translate” the alternative
nomenclature sometimes used in papers into this primary. As described in chapters three and
four on methods and results, an “alty” field in the ‘yDNA’ table records common alternative
names for ISOGG clade or subclade names. A field was eventually added to the
‘archaeological_find’ table called “ydna_notes” to record any additional details that potentially
could be used to update the table if advances or changes in the genetic research call for a revision
of the geodatabase tables to allow for better queries concerning Y-DNA clades.
Future work for this project involves a continuation of data collection and expansion of
the geodatabase. Because of the likelihood that some potential users would be excluded because
of the licensing costs required to create and maintain an Esri enterprise geodatabase needed to

55

allow for certain sharing and access controls, open source alternatives should be explored. This
will be discussed later in the chapter.
5.2. Data Gaps
The greater availability of samples from certain regions is partly because of their climate
or environment, such as being found in bogs, and consequently harboring better suitability for
DNA analysis. There is now a greater amount of work being carried out in regions of Asia (Fu et
al. 2014; 2013), Siberia, and present-day China, for example, than what might be inferred from
the papers available for incorporation into the geodatabase. Most papers are published in English
to share internationally, but by searching academic work published in French, Mandarin, or
Russian, it may be possible to increase the amount of usable data. Some of the difference in
availability of samples from certain regions also results from cultural and legal circumstances.
Advances in technology may close the gap due to challenges processing samples from some
environments, such as the growing availability of DNA extracted from archaeological finds near
the tropics (Slatkin and Racimo 2016). However, the cultural and legal reasons that impede, or in
some cases completely prohibit, work on human remains are unlikely to change soon.
5.2.1. Acknowledging the Data Gap
Although samples from African and Pacific regions with the necessary DNA analysis
were found throughout the course of the project, the data remained heavily skewed toward
northern regions of Asia and Europe. North American sample sizes were also not as abundant.
Africa is crucial for understanding the earliest human ancestors and their migrations (Pääbo
2014), yet many samples with published analysis are relatively recent compared to the contents
of the geodatabase. Without an acceptable sample size that includes a full spatio-temporal range,

56

it is not possible to state with confidence that fossil DNA either confirms or contradicts the work
of genetic analysis in resolving many questions of human migration and evolution.
As an example of the data gap problem, one of the sample queries illustrated in figure 13,
on page 57, shows the maternal DNA or mtDNA clades belonging to haplogroup M. The earliest
appearing clades on the map are found in two regions of Africa and southeastern Europe, close to
Asia Minor. This suggests a picture somewhat at odds with other research on haplogroup M’s
early branching off and its earliest dispersal (Metspalu 2004). This research explores today’s
Asian distribution of the haplogroup and posits that M haplogroup may have arisen in
southwestern Asia, although Metspalu (2004) admits possibility of an origin in east Africa. The
analysis based on this geodatabase appears to support an east African origin for this haplogroup,
but this is weakened as it must be acknowledged that better representation in the data for
southern Asia might reveal evidence for even earlier appearances there than east Africa.
Although not ideal, the sample sizes for Africa, the Levant, most of Southwest Asia, and
Central Asia are acceptable for some analysis of specific clades or larger haplogroups, and
investigation into spatio-temporal patterns. The samples from the Americas also allow for some
investigation. However, the lack of available samples from Pacific regions, especially Australia,
only allows very limited analysis at this stage. The ability to examine global spatio-temporal
patterns and fully investigate movements of genetic markers across the world will require better
representation in some regions.
5.2.2. Technology and Global Sample Availability
Since being founded more than two decades ago, the field of ancient DNA research
continues to grow but shares a common problem with forensics, that the amount of DNA
available within the samples is often limited (Rohland and Hofreiter 2007). As discussed in

57

Chapter 3, the techniques used to extract and analyze human DNA have steadily improved over
the last twenty years, so that samples found in climates that were once not ideal for this analysis
now may be processed successfully. Sampling from tropical locations has lagged behind
temperate and arctic regions partly due to better preservation of ancient DNA, but with recent
excavations such as Mota Cave in Ethiopia, perhaps more revealing discoveries will emerge
from Oceania and Africa (Slatkin and Racimo 2016).
With time, researchers may be able to run analysis on many more finds that previously
had not been categorized for mtDNA or Y-DNA haplogroups due to difficulty in extraction of
uncontaminated DNA with the required quality and quantity. For example, two famous
individuals found in Australia near Lake Mungo were omitted from this geodatabase due to
controversies over the proper identification of their DNA lineages. DNA analysis on the finds
was revisited recently (Heupink et al. 2016), and although these results were not included at this
time, it is likely that these later identifications of mtDNA or Y-DNA will be found to be reliable.
This hopefully will lead to more samples being conclusively typed and yield much-needed
diversity in the collection of analyzed samples. Continued efforts to collect data on
archaeological finds and incorporate them into a geodatabase will create a constantly improving
resource for investigation of archaeogenetics and all the fields making use of it.
5.2.3. Cultural, Legal, and Moral Considerations
The size of available analysis to incorporate can be expected to grow in most regions, but
in some the handling of human remains is subject to strict limitations. Although ethics figure
largely in any scientific endeavor involving human remains, the cultural and religious practices
of some peoples in the Americas and Australia control what is permissible and what is not in the
treatment of finds that could have belonged to an ancestor. Laws and cultural practices do not

58

necessarily prohibit the study of human remains; nevertheless, some institutions may not
participate in this because of the legal requirements and sensitivity. It is highly likely that the
size of samples from these continents will remain smaller in proportion to those from others.
In the U.S., the Native American Graves Protection and Repatriation Act (NAGPRA)
holds force. The software for databasing and managing museum holdings, EMu by Axiell,
includes tools for North American users to comply with legal requirements linked to indigenous
people’s artifacts and physical remains, such as repatriation after allowed studies as specified in
NAGPRA (“EMu – Collections Trust” n.d.; “Axiell Go” n.d.). Australian public bodies are
similarly bound to comply with the Return of Indigenous Cultural Property Program (Feikert
2009).
Although these laws and those of other nations with indigenous populations may not
necessarily prevent any scientific examination of remains, scientific testing may be delayed or
outright prohibited while authorities determine if the remains belong under control of a certain
group. The beliefs, customs, and leadership of the group deemed to have authority will determine
the outcome. In some cases, this means reburial (or other cultural funerary practice) by the group
once analysis is done, but in others, DNA analysis would not be permitted whatsoever. For
various reasons, some organizations such as the Australian Capital Territory (Feikert 2009)
choose not to hold artifacts or remains at all, and so do not participate in the legal guidelines.
The famous case of the “Kennewick Man” found in 1996 by teenagers in Washington
state, is one such instance. Initially, some researchers believed that the remains were possibly
related to other ancient peoples from Eurasia that had made the journey to North America in
prehistoric times, but later disappeared as a distinctive group - or perhaps left no descendants at
all. As such, claims to the remains by the tribes of Columbia Plateau and Nez Perce were not

59

recognized by scientists (Goldberg 2006). However, the Army Corps of Engineers halted testing,
agreeing with the Native American tribes, and a court case followed. Numerous appeals came to
the courts, with the right to scientific access finally being hinged upon the need for proof that
“human remains bear some relationship to a presently existing tribe, people, or culture...” as the
US Court of Appeals for the Ninth Circuit explained in Bonnichsen vs. United States (2004). The
tribes declined to take the case to the Supreme Court, and instead chose to focus their efforts on
making changes to NAGPRA.
Whether because of the legal burdens or respect for the spiritual beliefs and emotions
involved, it is likely that many researchers will not seek to work on human remains where it
would not be acceptable to the peoples who believe themselves to be their descendants or
relations. Aside from these present-day ethical and legal considerations, other moral
considerations related to past abuses certainly impact international efforts to examine human
remains. Disagreements over ownership and proper care of artifacts, mummies, and other
remains abound, often a result of colonial practices or flagrant disregard for local law or
sensitivities by past collectors.
To protect their cultural heritage and ensure full access to finds for their own researchers,
many institutions, particularly those in former colonies (Porr and Matthews 2019), may resist
loaning remains to foreign laboratories, especially for DNA extraction. The process requires
destruction of at least a small part of the remains, which both damages the find and limits the
number of times it can be repeated. However, not all institutions are equipped to carry out this
testing, and so work on some samples may be delayed.
These reservations about excavations and analysis on human remains in many areas are
complex and understandable (Porr and Matthews 2019). They are unlikely to be resolved to the

60

satisfaction of all with a stake in the future of human archaeological finds. For this reason, the
availability of a fully representative sample of ancient humanity will not be realized soon,
although improved techniques in retrieval from archaeological finds open to investigation will
bring it much closer to fruition.
5.3. Lessons Learned
Overall, working in Esri’s geodatabase format using ArcGIS Pro was uncomplicated and
suitable for the needs of the project. Data management tasks, such as creating and working with
domains went well. Writing and running queries was not difficult, although some minor issues
were noted. The file geodatabase remained a very appropriate choice for work with a project at
the present size with one individual responsible for entries and edits.
5.3.1. Domains
Working with domains in the ‘archeological_find’ helped ensure that values were
appropriate and made work with the related DNA tables go smoothly. For this project, domains
keep the values of mtDNA and Y-DNA clades to recognized formats. Because so many clade
and sub-clade names were possible in this geodatabase, the domains were auto-generated using a
readily available script, Esri ArcGIS Pro’s ‘Table-to-Domain’ tool. The source tables were
genetic haplogroups tables created to form the basis of the Y-DNA and mtDNA tables in the
geodatabase, compiled from names and relationships of the clades and haplogroups described in
references for genetic genealogy (“MtDNA – Results (MtDNA – Mutations) – FamilyTreeDNA
Learning Center” n.d.) and DNA resource pages (“Welcome to ISOGG... | International Society
of Genetic Genealogy” n.d.).
To reduce errors being introduced at the very beginning, the DNA tables to be used as the
foundation of the domains in the ‘archaeological_find’ table were checked for unique values

61

with filtering options in Excel. Next, this column of clades was double-checked for any entries
that did not match known conventions or formats. Once the verification was complete, the tool
was used to create domains for Y-DNA and mtDNA clades in the ‘archaeological_find’ table.
The large number of coded values made the drop-down menu quite long. Nevertheless, it
was not cumbersome since the GIS could locate the section of the drop-down when the first
characters were entered. The use of numbered types with domains was considered in order to
shorten these lists, since the DNA sub-clades, clades and sub-haplogroups could be organized by
the older upstream group. However, the use of numbers to represent more than four or five
groups did not seem to be a user-friendly solution; it would not be reasonable to expect someone
to mentally recall correspondences for so many values or refer to external notes. Continuing with
the large number of coded values with their widely understood alphanumeric system seemed
preferable. If Esri ever offers subtypes using other codes or a built-in lookup option, this could
be implemented without frustrating an editor or possibly introducing more error.
5.3.2. Tool Use in the File Geodatabase
Some tools and features were lacking for the file geodatabase. For some functions, an
enterprise geodatabase is required. The cost of the enterprise geodatabase license was
prohibitive, and the file geodatabase was adequate for the purpose of the project at this stage.
One example of unavailable tools is the “Make Query Layer” tool. The “Make Table Query” tool
was available and provides comparable results. The geometry field can be included in “Make
Query Table” tool to give a mappable result or can be omitted to give data tables only.
The relationship classes in Esri worked as expected, allowing for normalized tables. The
only drawback to this was that executing some queries in ArcGIS Pro was less straightforward
than they would have been using DBMS tools. One issue encountered was inconsistent behavior

62

in one tool using SQL. The SQL window in the “Make Query Table” tool has an option to verify
that the SQL entered is valid. In at least one instance, the SQL was returned as “valid”, yet the
query failed, giving the “An invalid SQL statement was used” error. Conflicting messages as
these do not help establish whether an alternative form of SQL may be used, if the user did not
construct the query properly, or some other issue was at fault. Nevertheless, by using the
aforementioned “Query Table” or by using simple spatial joins and/or definition, query results
could be obtained for queries requiring data stored in the separate tables.
5.4. Future Possibilities
Use of the file geodatabase format was simple and more affordable than an Esri
enterprise geodatabase. As discussed in 5.3, it was appropriate for its purpose. Nevertheless, for
a hypothetical future project involving multiple data owners and a much larger volume of data,
some options of the enterprise geodatabase might be desired. There are greater options for use of
SQL in direct data management, which is not recommended with the file geodatabase at present,
according to Esri’s help pages (“Supported Databases—ArcGIS Pro | Documentation” n.d.). As
mentioned in 5.3.2, not all tools are available in the file geodatabase. Enterprise would also allow
for multiple users to access and edit using versioning. As the project grows, a move to enterprise
might be more attractive.
5.4.1. Open Source Options?
The costs of Esri licensing can be prohibitive for many projects, and enterprise licensing
adds further expense. Esri offers many benefits, such as a high standard of visualization, a
multitude of tools and algorithms readily available for analysis of all kinds, curated data, and
resources for training and troubleshooting. This certainly justifies the costs for many
organizations, but all budgets do not allow an Esri option no matter how desirable.

63

5.4.1.1. Quantum GIS – QGIS 3.x
Use of the geodatabase in its present form does not require expensive licensing. For most
non-commercial purposes, there is a $100 fee for student or personal use license that can be
easily obtained from Esri. Even this minimal fee is not necessary to use the current
archaeogenetic database. Access to the file geodatabase in its current form is possible with recent
Quantum GIS (QGIS) versions, certainly all after 3.0 can read .gdb files. QGIS is an open source
project and is completely free to download and use with the full suite of tools that have been
developed for it. QGIS has added more training and user manual information to its website since
its initial appearance, and multiple videos on usage for new users can be found on YouTube free
of charge. QGIS also can work with a PostgreSQL database coupled with the PostGIS extension.
This offers some choices for porting the project to open source in the future if desired.
5.4.1.2. PostgreSQL
If use of an enterprise option should become desirable, but Esri licensing is still not
possible, then PostgreSQL with PostGIS for spatial capabilities could be a viable alternative.
FME Safe Software offers a solution to help convert a geodatabase to PostgreSQL, but this is
also a commercial solution. For an open source solution making use of GDAL, the code:
ogr2ogr -f "PostgreSQL" PG:"dbname=mydbname user=postgres" myFileGDB.gdb
can be used. User Burham also noted that “FileGDB” had to be installed for this to work, but
“ogrinfo –formats” could be used to verify its installation (“Ogr2ogr - How to Import ESRI
Geodatabase Format .Gdb into PostGIS” n.d.).
As PostgreSQL with PostGIS has continuously updated its offerings and released new
versions since 2012, it is likely that this solution, or a similar one, can still work with file
geodatabases created with more recent ArcGIS Pro versions. As of 2020, a package available on

64

GitHub (Iannou 2021) has been used to convert the full file geodatabase, including domains, for
use with PostGIS.
Other open source databases with spatial capabilities exist, so further investigation into
these options and the benefits or problems associated with them would be needed before
migrating the geodatabase (Badea and Badea 2018). Both users of Esri and of QGIS can connect
to these alternatives, as well as use the present geodatabase format. Should the project achieve
more success and require better options for easy sharing or other needs than the file geodatabase
allows, these open source enterprise database solutions are promising. However, work in the file
geodatabase format in this project has allowed for simple data management, along with a highly
convenient integration with a top-notch GIS and all that it offers.
In conclusion, building the archaeogenetic geodatabase using Esri’s ArcGIS Pro was
successful. The geodatabase structure was able to easily execute queries like those made in a
prototype relational database. Data management tools worked well, which remains important as
the project grows with the publication of more analyzed DNA samples. The spatial capability
allows for the easy incorporation of special newly published spatial datasets, such as vector files
of historic geographic features. Should the project eventually require a move away from Esri
software, there are options to migrate the geodatabase without requiring it to be rebuilt. The
archaeogenetic geodatabase has proven to be an expandable data repository for spatial analysis.

65

References

Achilli, Alessandro, Ugo A. Perego, Hovirag Lancioni, Anna Olivieri, Francesca Gandini,
Baharak Hooshiar Kashani, Vincenza Battaglia, et al. 2013. “Reconciling Migration
Models to the Americas with the Variation of North American Native Mitogenomes.”
Proceedings of the National Academy of Sciences 110 (35): 14308–13.
https://doi.org/10.1073/pnas.1306290110
Arias, Leonardo, Roland Schröder, Alexander Hübner, Guillermo Barreto, Mark Stoneking, and
Brigitte Pakendorf. 2018. “Cultural Innovations Influence Patterns of Genetic Diversity
in Northwestern Amazonia.” Edited by Connie Mulligan. Molecular Biology and
Evolution, August. https://doi.org/10.1093/molbev/msy169.
“Axiell Go.” n.d. Accessed October 24, 2020.
http://help.emu.axiell.com/v6.0/en/Topics/EMu/Axiell%20Go.htm.
Badea, Ana, and Gheorghe Badea. 2018. CONSIDERATIONS ON OPEN SOURCE GIS
SOFTWARE VS. PROPRIETARY GIS SOFTWARE.
Baumann, Bailey. n.d. “Finding Environmental Opportunities for Early Sea Crossings: An
Agent-Based Model of Middle to Late Pleistocene Mediterranean Coastal Migration,” 69.
Beyin, Amanuel. 2011. “Upper Pleistocene Human Dispersals out of Africa: A Review of the
Current State of the Debate.” International Journal of Evolutionary Biology 2011.
Burrows, Cynthia. n.d. “Developing an Archaeological Specific Geodatabase to Chronicle
Historical Perspectives at Bethsaida, Israel,” 90.
Cavalli-Sforza, Luigi L. and Cavalli-Sforza. 1994. “THE GREAT HUMAN DIASPORAS: A
HISTORY OF DIVERSITY AND EVOLUTION.” 136578. 1994.
https://repository.library.georgetown.edu/handle/10822/545601.
“Converting GIS Vector Data to KML | Keyhole Markup Language.” n.d. Google Developers.
Accessed March 22, 2021. https://developers.google.com/kml/articles/vector.
“EMu – Collections Trust.” n.d. Accessed October 24, 2020.
https://collectionstrust.org.uk/software/emu/.
Feikert, Clare. 2009. “Repatriation of Historic Human Remains: Australia.” Web page. July
2009. https://www.loc.gov/law/help/repatriation-human-remains/australia.php.
Fu, Qiaomei, Heng Li, Priya Moorjani, Flora Jay, Sergey M. Slepchenko, Aleksei A. Bondarev,
Philip L. F. Johnson, et al. 2014. “Genome Sequence of a 45,000-Year-Old Modern
Human from Western Siberia.” Nature 514 (7523): 445–49.
https://doi.org/10.1038/nature13810.

66

Fu, Qiaomei, Matthias Meyer, Xing Gao, Udo Stenzel, Hernán A. Burbano, Janet Kelso, and
Svante Pääbo. 2013. “DNA Analysis of an Early Modern Human from Tianyuan Cave,
China.” Proceedings of the National Academy of Sciences 110 (6): 2223.
https://doi.org/10.1073/pnas.1221359110.
Gimbutas, Marija. 1963. “The Indo-Europeans: Archeological Problems.” American
Anthropologist 65 (4): 815–36.
Grugni, Viola, Alessandro Raveane, Linda Ongaro, Vincenza Battaglia, Beniamino Trombetta,
Giulia Colombo, Marco Rosario Capodiferro, et al. 2019. “Analysis of the Human Y-
Chromosome Haplogroup Q Characterizes Ancient Population Movements in Eurasia
and the Americas.” BMC Biology 17 (1): 3. https://doi.org/10.1186/s12915-018-0622-4.
Haak, Wolfgang, Guido Brandt, Hylke N. de Jong, Christian Meyer, Robert Ganslmeier, Volker
Heyd, Chris Hawkesworth, Alistair W. G. Pike, Harald Meller, and Kurt W. Alt. 2008.
“Ancient DNA, Strontium Isotopes, and Osteological Analyses Shed Light on Social and
Kinship Organization of the Later Stone Age.” Proceedings of the National Academy of
Sciences 105 (47): 18226–31. https://doi.org/10.1073/pnas.0807592105.
Heupink, Tim H., Sankar Subramanian, Joanne L. Wright, Phillip Endicott, Michael Carrington
Westaway, Leon Huynen, Walther Parson, Craig D. Millar, Eske Willerslev, and David
M. Lambert. 2016. “Ancient MtDNA Sequences from the First Australians Revisited.”
Proceedings of the National Academy of Sciences 113 (25): 6892–97.
https://doi.org/10.1073/pnas.1521066113.
Hrdlička, A. 1907. Skeletal Remains Suggesting or Attributing to Early Man in North America.
Washington, DC: Government Printing Office.
Hrdlicka, A. 1936. The Coming of Man from Asia in the Light of Recent Discoveries.
Washington, DC: Smithsonian Institute.
Hublin, Jean-Jacques, Abdelouahed Ben-Ncer, Shara E. Bailey, Sarah E. Freidline, Simon
Neubauer, Matthew M. Skinner, Inga Bergmann, Adeline Le Cabec, and Stefano
Benazzi. 2017. “New Fossils from Jebel Irhoud, Morocco and the Pan-African Origin of
Homo Sapiens.” Nature Jun 7;546(7657):289-292. https://doi.org/10.1038/nature22336.
Iannou, GM. (2017) 2021. Cartologic/Fgdb2postgis. Python. Cartologic.
https://github.com/cartologic/fgdb2postgis.
Krzewińska, Maja, Gülşah Merve Kılınç, Anna Juras, Dilek Koptekin, Maciej Chyleński, Alexey
G. Nikitin, Nikolai Shcherbakov, et al. 2018. “Ancient Genomes Suggest the Eastern
Pontic-Caspian Steppe as the Source of Western Iron Age Nomads.” Science Advances 4
(10): eaat4457. https://doi.org/10.1126/sciadv.aat4457.
Morrish, Seán William, and Debra F. Laefer. 2010. “Web-Enabling of Architectural Heritage
Inventories.” International Journal of Architectural Heritage 4 (1): 16-37.
https://doi.org/10.1080/15583050902731056.

67

“MtDNA – Results (MtDNA – Mutations) – FamilyTreeDNA Learning Center.” n.d. Accessed
February 6, 2021. https://learn.familytreedna.com/user-guide/mtdna-myftdna/mt-results-
page/.
“Ogr2ogr - How to Import ESRI Geodatabase Format .Gdb into PostGIS.” n.d. Geographic
Information Systems Stack Exchange. Accessed February 27, 2021.
https://gis.stackexchange.com/questions/83016/how-to-import-esri-geodatabase-format-
gdb-into-postgis.
“Opening File Based Geodatabases in QGIS 2.4 • North River Geographic Systems Inc.” 2014.
North River Geographic Systems Inc (blog). September 8, 2014.
https://www.northrivergeographic.com/archives/opening-file-based-geodatabases-qgis-2-
4.
Pääbo, Svante. 2014. Neanderthal Man: In Search of Lost Genomes. Basic Books.
Porr, Martin, and Jacqueline Matthews. 2019. Interrogating Human Origins: Decolonisation and
the Deep Human Past. Routledge.
Rasheed, Haroon ur, Muhammad Jawad, Shahid Nazir, Saadia Noreen, and Allah Rakha. 2017.
“MtDNAmap: Geographic Representation of MtDNA Haplogroups.” Forensic Science
International: Genetics Supplement Series 6 (December): e516–17.
https://doi.org/10.1016/j.fsigss.2017.09.208.
Reich, David, Nick Patterson, Desmond Campbell, Arti Tandon, Stéphane Mazieres, Nicolas
Ray, Maria V. Parra, et al. 2012. “Reconstructing Native American Population History.”
Nature 488 (7411): 370–74. https://doi.org/10.1038/nature11258.
“RFC 7946 - The GeoJSON Format.” n.d. Accessed March 22, 2021.
https://tools.ietf.org/html/rfc7946.
Rohland, Nadin, and Michael Hofreiter. 2007. “Comparison and Optimization of Ancient DNA
Extraction.” BioTechniques 42 (3): 343–52. https://doi.org/10.2144/000112383.
Slatkin, Montgomery, and Fernando Racimo. 2016. “Ancient DNA and Human History.”
Proceedings of the National Academy of Sciences 113 (23): 6380–87.
https://doi.org/10.1073/pnas.1524306113.
Stringer, C.B., and P. Andrews. 1988. “Genetic and Fossil Evidence for the Origin of Modern
Humans.” Science 239 (4845): 1263–68.
Stringer, Chris, and Peter Andrews. 2005. The Complete World of Human Evolution. London &
New York: Thames & Hudson.
“Supported Databases—ArcGIS Pro | Documentation.” n.d. Accessed September 21, 2020.
https://pro.arcgis.com/en/pro-app/help/data/databases/dbms-support.htm.

68

Twumasi, Bright Osei. 2002. “Modelling Spatial Object Behaviours in Object- Relational
Geodatabase.” International Institute for Geo-Information Science and Earth Observation.
Underhill, Peter A., and Toomas Kivisild. 2007. “Use of Y Chromosome and Mitochondrial
DNA Population Structure in Tracing Human Migrations.” Annual Review of Genetics 41
(1): 539–64.
Underhill, Peter A., G. David Poznik, Siiri Rootsi, Mari Järve, Alice A. Lin, Jianbin Wang, Ben
Passarelli, et al. 2015. “The Phylogenetic and Geographic Structure of Y-Chromosome
Haplogroup R1a.” European Journal of Human Genetics 23 (1): 124–31.
https://doi.org/10.1038/ejhg.2014.50.
Viciani, D., F. Geri, N. Agostini, V. Gonnelli, and L. Lastrucci. 2018. “Role of a Geodatabase to
Assess the Distribution of Plants of Conservation Interest in a Large Protected Area: A
Case Study for a Major National Park in Italy, Plant Biosystems.” An International
Journal Dealing with All Aspects of Plant Biology 152 (4): 631-641,.
https://doi.org/10.1080/11263504.2017.1308974.
Wei, Lan-Hai, Ling-Xiang Wang, Shao-Qing Wen, Shi Yan, Rebekah Canada, Vladimir
Gurianov, Yun-Zhi Huang, et al. 2018. “Paternal Origin of Paleo-Indians in Siberia:
Insights from Y-Chromosome Sequences.” European Journal of Human Genetics 26
(11): 1687–96. https://doi.org/10.1038/s41431-018-0211-6.
“Welcome to ISOGG... | International Society of Genetic Genealogy.” n.d. Accessed February 6,
2021. https://isogg.org/.
Wilson, Allan C., and Rebecca L. Cann. 1992. “The Recent African Genesis of Humans.”
Scientific American 266 (4): 68–75.
“Working with File Geodatabases (.Gdb) Using QGIS and GDAL | Geospatial @ UCLA.” n.d.
Accessed October 25, 2020. https://gis.ucla.edu/node/53.
Yang, Ning Ning, Stéphane Mazières, Claudio Bravi, Nicolas Ray, Sijia Wang, Mari-Wyn
Burley, Gabriel Bedoya, et al. 2010. “Contrasting Patterns of Nuclear and MtDNA
Diversity in Native American Populations.” Annals of Human Genetics / Stories of
Human Genetics, September, 74(6):525-38. https://doi.org/10.1111/j.1469-
1809.2010.00608.x

69
Appendix A Data Table

Description of the data sets and tables
region
Low resolution -
global/Extent:West
-179.999989
East 179.999989
North 83.623600
South -89.000000 Holocene (current epoch)
Esri shapefile =
polygon Yes Esri data portal
mtDNA Nonspatial
from early as 100 kya to
present table, nonspatial
Yes, but
expanding
Created through Family Tree DNA,
ISOGG
yDNA Nonspatial
from early as 100 kya to
present table, nonspatial
Yes, but
expanding
Created through Family Tree DNA,
ISOGG
archaeological_find
Moderate
resolution-
Coordinate
precision to 6
decimal points
(using location
names in
sources)/ Extent
West -
175.110590
East 170.230494
North 70.776562
South -55.248573 43,000 BCE to 1450 AD
Esri shapefile
= point Yes
Multiple published papers-
including Mathiesen et al. 2017;
OSM public project served as reference
to locate relevant papers quickly

70
Appendix B Published Data

Excel Table included in Mathiesen et al. (2017) as supplemental material

Analysis_Label Culture Sample_ID Y-HG mtDNA Average_ Date Location Country Sex Coverage SNPs
Balkans_BronzeAge Bulgaria_Ezero_EBA Bul10 .. .. 4957 3090-2924 calBCE Sabrano Bulgaria F 0.066 66247
Balkans_BronzeAge Bulgaria_Beli_Breyag_EBA Bul6 I2a2 .. 4450 3400-1600 BCE Beli Breyag Bulgaria M 0.823 370439
Balkans_BronzeAge Bulgaria_Beli_Breyag_EBA Bul8 I .. 4450 3400-1600 BCE Beli Breyag Bulgaria M 0.017 18337
Balkans_BronzeAge Bulgaria_MLBA I2163 R1a1a1b2 U5a2 3638 1750-1625 calBCE Merichleri, K Bulgaria M 4.106 825494
Balkans_BronzeAge Bulgaria_EBA I2165 I2a2a1b1b T2f 4908 3020-2895 calBCE Merichleri, K Bulgaria M 5.55 857708
Balkans_BronzeAge Bulgaria_EBA I2175 I2a2a1b1 K1c1 5122 3328-3015 calBCE Smyadovo Bulgaria M 0.527 423781
Balkans_BronzeAge Bulgaria_BA I2510 G2a2a1a2 H4a1 4758 2906-2710 calBCE DzhulyunitsaBulgaria M 6.7 821681
Balkans_BronzeAge Bulgaria_BA I2520 H2 H 5132 3336-3028 calBCE DzhulyunitsaBulgaria M 6.117 795071

71
Image of data as published by Prendergast et al. 2019

72
Appendix C Domains
List of mtDNA domain coded values
A2c C1c D1t G3a H28 H45 H7a I3a K2b M13 M5b Q1b T2 U4 V14 X2b
A2d C1d D2a G3b H29 H46 H7b I4a K3 M1a M65 Q2a T2+ U4a V17 X2c
A2h C1e D4 H18 H2a H49 H7c I5a L0a M1b M70 R T2a U4b V1a X2d
A2i C1g D4a H1a H2b H4a H7d I5b L0d M20 M7b R+1 T2b U4c V1b X2f
A2p C4 D4b H1b H2c H4d H7f I6 L0f M21 M7c R0 T2c U4d V3 X2i
A4f C4a D4e H1c H3 H5 H8c J L0k M28 M8a R0a T2d U5 V7a X2l
A8a C4b D4h H1e H3+ H5' H92 J1 L1c M3 M9a R1a T2e U5a W X2m
B C4d D4j H1f H30 H5+ H9a J1 L2a M30 N R1b T2f U5b W1 X2p
B2 C5c D4m H1g H32 H5a HV J1+ L3b M33 N1a R2 T2g U6a W1- X4
B2a C7a D4o H1h H33 H5b HV- J1b L3d M35 N1b R2+ T2h U6b W1+ Z1
B2b Cb1 D4q H1i H35 H5c HV+ J1c L3e M3a N9a R3 T2k U6d W1c Z1a
B2i

D5a H1j H3a H5d HV0 J1d L3f M3c N9b R30 U* U7a W1e Z3a
B2y

F1a H1k H3b H5n HV1 J2a L3h M4

R5a U1a U7b W3a

B4a

F1b H1n H3c H60 HV2 J2b L3i M49

R6a U1b U8a W3b

B5a

F1d H1q H3f H65 HV4 K L3x M4a

R6b U2 U8b W5

B5b

F1e H1t H3g H66 HV6 K1 L4a

R7 U2' U8c W5a

F1f H1u H3h H67 HV9 K1 L4b

S2a U2+

W6

F2a H2 H3t H6a I1 K1a L5b

U2a

W6a

F2c H2+ H3u H6b I1a K1b

U2b

W6c

F2g H23 H3v H6c I1b K1c

U2c

I1c K1d

U2d

U2e

73
List of Y-DNA domain coded values
A00 E G2a2b I I2a2 J L Q R1a
A1 E1b1b1b2 G2a2b1 I1 I2a2a J1 L1 Q1 R1a1
A1b1b2 E1b1b1b2a G2a2b2a I1a1b1 I2a2a1 J1a L1a Q1a R1a1’2
A1b1b2a E1b1b1b2b2a1 G2a2b2a1a I1a1b3 I2a2a1 J1a2a1a2d2b2 L2 Q1a* R1a1a
A1b1b2b E1bE1b1b1a1a1b1 G2a2b2a1a1b1 I1a1b3b I2a2a1a1a J1a2b M1b Q1a1 R1a1a1
B E2 G2a2b2a1a1b1a1a1 I1a2a1a2 I2a2a1a1a1 J2 N Q1a1b R1a1a1?
B2b F G2a2b2a1a1c1a I1a3 I2a2a1a1a2 J2a N1a Q1a1b1 R1a1a1b
BCDEF F* G2a2b2a1c I1b I2a2a1a2a1a J2a1 N1a1a1a1a4a1 Q1a2 R1a1a1b1a2
xBT G G2a2b2a3 I2 I2a2a1b J2a1a2a2 N1c1a Q1a2a R1a1a1b1a2b1
BT G1a G2a2b2b I2a I2a2a1b1 J2a1d N1c1a1a Q1a2a1 R1a1a1b1a3
C G2 G2a2b2b1 I2a1 I2a2a1b1b J2a1h N1c2 Q1a2a1a R1a1a1b1a3a
C1a G2a G2a2b2b1a I2a1a1 I2a2a1b1b1 J2a1h2 N3a3′5 Q1a2a1a1 R1a1a1b1a3a1
C1a2 G2a1 G2a2b2b1a1 I2a1a1a I2a2a1b2 J2a2a N3a3a Q1a2a1a1 R1a1a1b1a3b
С1b G2a1a1 G2a2b2b1a1a I2a1a1a1a1a1a1e5~ I2a2a1b2a2 J2a8 NO Q1a2a1b R1a1a1b2
C1a2a G2a2 G2b I2a1a1a1b I2a2a1b2a2a2 J2b O Q1a2a1c R1a1a1b2a
C1b1a1a1 G2a2a H I2a1a2 I2a2a2 J2b2a O1a Q1a2b R1a1a1b2a2a
C2b G2a2a1 H1a1 I2a1a2a1a I2a2a2a J2b2a1 O1a1a1a Q1a2b2 R1a1a1b2a2b
C2b1a1 G2a2a1a H1a1a I2a1b I2b K O1b Q1aa1c R1a1c
C2b1a1a G2a2a1a2 H1a1d2 I2a1b1 I2c K2 O1b1a1a1b Q1b1 R1a5
C2b1a1b2 G2a2a1a2a H1a2a1 I2a1b1a I2c1 K2b1 O1b1a1a1b1 Q1b2 R1b
C3 G2a2a1a2a1 H1b1 I2a1b1a1 I2c2 K2b1a3 O2a Q1c

C6 G2a2a1a2a1a H2 I2a1b2 I2d K2b2a2 O2a1c1b1a R

CT G2a2a1a3 H3b

IJ

O3a R*

D G2a2a1b HIJ

IJK

O3a3b2 R1

D1b2b G2a2a1b1

P1 R1*

74
List of Y-DNA coded value domains
R1b1 R1b1a1a2a1a2d R1b1a2a2
R1b1a R1b1a1a2a1a2f R1b1a2a2c1
R1b1a1 R1b1a1a2a2 R1b1b
R1b1a1a R1b1a1a2a2c1 R1b1b2
R1b1a1a1 R1b1a1b R1b1b2a
R1b1a1a2 R1b1a1b1 R1b1b2g
R1b1a1a2a R1b1a2 R2
R1b1a1a2a1 R1b1a2a R2a
R1b1a1a2a1a R1b1a2a1a R2a3a
R1b1a1a2a1a1 R1b1a2a1a1 R2a3a2
R1b1a1a2a1a1b R1b1a2a1a1b R2a3a2b
R1b1a1a2a1a1c R1b1a2a1a1c R2a3a2b2b1
R1b1a1a2a1a1c1a R1b1a2a1a1c2b2b R2a3a2b2c
R1b1a1a2a1a1c2b2a1b1a R1b1a2a1a1c2b2b1a1 S1a
R1b1a1a2a1a1c2b2b1a1a1 R1b1a2a1a2 T
R1b1a1a2a1a2 R1b1a2a1a2* T1a
R1b1a1a2a1a2a1 R1b1a2a1a2b T1a1
R1b1a1a2a1a2a1b R1b1a2a1a2c T1a1a
R1b1a1a2a1a2a5 R1b1a2a1a2c1 T1a1a1b2
R1b1a1a2a1a2b1 R1b1a2a1a2c1g T1a2b
R1b1a1a2a1a2c R1b1a2a1a2c1g1a1

R1b1a1a2a1a2c1 R1b1a2a1a2c2

R1b1a1a2a1a2c1a1

R1b1a1a2a1a2c1e2b3

R1b1a1a2a1a2c1e2b3a1

Abstract (if available)

Abstract From its early beginnings from the parent genus Homo in Africa, the species Homo sapiens spread across the globe to every continent except Antarctica, long before the advent of large seafaring vessels or even the wheel. The dispersion of the first Homo sapiens occurred when other early human species, such as Neanderthals or Denisovans, were still in Europe and parts of Asia, and land features and climates were very different from northern and eastern Africa. As early modern humans encountered these new environments and possibly other, earlier peoples over centuries of migration, adaptations occurred, and new cultures arose. These migrations are of great interest to several disciplines, including physical anthropology, archaeology, and genetics. A global geodatabase as a repository of spatial and genetic data to facilitate Spatio-temporal models of models and various data visualizations would serve all these disciplines. Such a geodatabase also can incorporate other related data for investigation, such as global regions, early coastlines, glacier limits, or the overall continental geography of earlier ages for investigation of their possible effects on movement or settlement of ancient peoples. Additionally, a geodatabase offers many options to share or limit access to data. This project offers a comprehensive data source and tool for creating and sharing analyses with other research efforts.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Web-based relational spatiotemporal geodatabase of glaciers in California

PDF

Surface rupture detection with support vector machine classification: case study from Ridgecrest, CA

PDF

Evaluating machine learning tools for humanitarian road network mapping

PDF

A unified geodatabase design for sinkhole inventories in the United States

PDF

The movement of Mexican migration and its impact based on a GIS geospatial database

PDF

Creating a water quality geodatabase for the West Hawai‘i Island region

PDF

GIS data curation and Web map application for La Brea Tar Pits fossil occurrences in Los Angeles, California

PDF

A fire insurance map geocoder for pre-earthquake San Francisco

PDF

Utilizing existing museum collections and GIS for paleontological site assessment and management

PDF

Operational optimization model for Hungry Marketplace using geographic information systems

PDF

Urban areas and avian diversity: using citizen collected data to explore green spaces

PDF

Developing an archaeological specific geodatabase to chronicle historical perspectives at Bethsaida, Israel

PDF

San Diego Wildfire Hazards Information Center mashup

PDF

Semi-automated visualization of spatial information in unstructured text

PDF

Finding your best-fit neighborhood: a Web GIS application for online residential property searches for Anchorage, Alaska

PDF

Geographic object based image analysis for utility scale photovoltaic site suitability studies

PDF

Generating bicyclist counts using volunteered and professional geographic information through a mobile application

PDF

GeoBAT: crowdsourcing dynamic perception of safety data through the integration of mobile GIS and ecological momentary assessments

PDF

Estimating at-risk population for lead service lines induced lead exposure and their correlation to socioeconomically disadvantaged neighborhoods in Milwaukee, Wisconsin

PDF

Visualizing email response data to improve marketing campaigns

Asset Metadata

Creator Leasure, Maria del Carmen (author)

Core Title Geodatabase for archaeogenetics: ancient peoples and family lines

School College of Letters, Arts and Sciences

Degree Master of Science

Degree Program Geographic Information Science and Technology

Degree Conferral Date 2021-08

Publication Date 07/18/2021

Defense Date 04/16/2021

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

Tag Anthropology,archaeogenetics,archaeology,early peoples,Genetics,geodatabase,OAI-PMH Harvest

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Bernstein, Jennifer Moore (committee chair), Chiang, Yao-Yi (committee member), Wu, An-Min (committee member)

Creator Email mleasure@usc.edu,moomolee@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-oUC15607944

Unique identifier UC15607944

Legacy Identifier etd-LeasureMar-9775

Document Type Thesis

Format application/pdf (imt)

Rights Leasure, Maria del Carmen

Type texts

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

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the author, as the original true and official version of the work, but does not grant the reader permission to use the work if the desired use is covered by copyright. It is the author, as rights holder, who must provide use permission if such use is covered by copyright. The original signature page accompanying the original submission of the work to the USC Libraries is retained by the USC Libraries and a copy of it may be obtained by authorized requesters contacting the repository e-mail address given.

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA

Repository Email cisadmin@lib.usc.edu