Inferring the Early History of
Northern South America

through Mitochondrial DNA
Analysis

Mateo Rojas, Miguel Vilar

 Abstract
Prior to the mass colonization of the Americas, early peopling consisted of Indigenous Americans who, over the course of tens
of thousands of years, came to occupy the continents that would later be settled by the ambitious European powers of the
time. Through the use of mitochondrial DNA (mtDNA) sequencing throughout numerous samples, as well as temporal analysis
of genetic mutations within samples of the same haplogroup, the migrations of different early indigenous populations can be
tracked across the geography of the continent, and the ethnographic composition of different regions can be reconstructed
with respect to the time period. Given that haplogroups belonging to maternal Amerindian descent are identified by
haplogroups A, B, C, and D, and that genetic variations in haplogroups can be tracked over time, it is possible to approximate a
model of the early genetic composition of Colombia and other regions of northern South America through the use of samples
from model haplogroups like A2al, B2d14, C1d2, and D4h3a. For this study, 94 samples across these haplogroups have been
surveyed, and individual phylogenetic trees have been constructed, which each infer a genetic timeline for the mutations
found under each haplogroup. With the construction of a phylogenetic tree for each of the four haplogroups, a larger,
cumulative phylogenetic tree was also constructed, which in combination with the known presences of haplogroups across the
geography of Colombia, can provide insight into the early migration and settling patterns of these early pre-columbian
Indigenous Americans.

Results
From our study, we managed to construct a unified phylogenetic tree and a migration pattern
for four unique Amerindian haplogroups, A2al, B2d14, C1d2, and D4h3a. We are looking
forward to expanding on this research in the future, as exploring not only other samples, but
the frequency of these haplogroups in regions of Central and South America will allow us to
determine more directionality in early migration patterns. And while our current sample size
of 94 is relatively small, it sets a precedent for how we and other researchers can continue to
fully apply the analysis of mtDNA in order to recreate the early settlements and peopling
across history. 

During the colonization of the Americas, indigenous populations were displaced, oppressed,
and exterminated, with the Native American populations during the expansion of the United
States struggling to integrate themselves into the new English-American status quo. Similarly,
indigenous populations in Central and South America were also uprooted to varying degrees.
In Central and Northern South America, people of Amerindian descent were either killed, or
absorbed into the Spanish and Portuguese populations, living on in their ethnic descendants,
but with their history completely obfuscated. Very few purely-indigenous populations remain
to this day, and are slowly dwindling. In Southern South America, settlers of present-day
Argentina, Chile, and Uruguay were known to be even more ruthless in their massacres of
indigenous populations, leaving little room even for ethnographic integration. Across the
Americas, there is very little indigenous history before colonization, at least compared to the
societies that once blossomed when left to their own devices.

In regards to our study, mtDNA analysis poses an invaluable avenue by which to study the
forgotten history of early Native American migrations and geographical diversity. It is
especially important to research the forgotten history of America’s early indigenous
populations, as well as other marginalized groups across history, as we are piecing back
together the lineages and stories of populations that still exist and flourish today. 

Conclusion 

Introduction
Ever since the first full sequencing of a human genome in
2002, DNA testing kits have become ubiquitous and highly
popular for those seeking genetic consultation, forensic
analysis, and for the most part, simple curiosity about one’s
ancestry and origin. Since then, companies such as Ancestry,
23andMe, and FamilyTreeDNA have grown to astonishing
proportions, and have compiled equally astonishing amounts
of data on the genetic history and composition of their
clientele. Utilizing FamilyTreeDNA’s online mtDNA database,
one is able to not only access anonymized samples,
haplogroups, and the individual mutations that comprise
them, but also geographic approximations for the dawn of
novel mutations and haplogroups across time. This allows
researchers to construct migratory patterns for populations of
different ethnographic origin, and the ability to simulate the
early peopling of different regions across the world.

Over the past few months, we have been utilizing this same
mtDNA database to compile the Amerindian genetic
composition of present-day Colombia, to simulate the
migratory patterns of these early-Amerindian lineages into
northern South America. 

Using the Time Estimates feature on
Network once the trees had been
constructed, and assuming an average
mutation rate of one mutation every 3600
years, we approximated the age of each
haplogroup (see Table 2).

And in combination with the migration
database on FamilyTreeDNA, we were able
to conclude, based on the data available,
how long it would have taken each
haplogroup to have migrated to their
present locations in Colombia from their
point of origin.

Haplogroup A2al, which differentiated from
haplogroup A in present-day Alaska, would
have taken around 13,000 years to migrate
down to South America.

Haplogroup B2d14, which differentiated
from haplogroup B in present-day Baranof
Island, would have taken around 7,000
years to migrate down.

Haplogroup C1d2, which differentiated from
haplogroup C in the present-day Amur
Oblast of Russia, would have taken around
11,000 years to migrate across to the
Americas and down.

And haplogroup D4h3a, which differentiated
from haplogroup D in the present-day
Northwestern Heilongjiang Province of
China, would have taken around 21,000
years to migrate across and down.

Mitochondrial DNA
mtDNA is present in almost all eukaryotes and multicellular
organisms in the form of haploidic, circular-shaped DNA
molecules, called plasmids. mtDNA, as opposed to nuclear
DNA, is inherited solely from one’s maternal lineage, and thus
is not exposed to the nature of genetic recombination. As a
result, mtDNA is very stable and consistent across time, while
still experiencing mutations around 10-20 times faster than
nuclear DNA. These comparatively rapid mutations coupled
with their preservation across time makes mtDNA highly
desirable for tracking one’s maternal ancestry, as the time
period and location in which novel mutations were acquired
can be approximated.

Figure 1. A depiction
of the unique

inheritance pattern
of mitochondrial

DNA, wherein only
maternal mtDNA is
inherited, and only

females pass it on to
their descendants.

Methodology
While haplogroups A, B, C and D are too large and historically long-lived to acquire specific enough results, especially solely by
human effort, using newer variants of these haplogroups that are prominent in samples acquired from Colombia are much
easier to analyze for research purposes. To this end, haplogroups A2al, B2d14, C1d2, and D4h3a were identified and served as
the primary focus for the study, given their presence in northern South America, and their workable sample sizes. A total of 94
samples were analyzed, comprised of 46 samples from haplogroup A2al, 15 samples from haplogroup B2d14, 10 samples from
haplogroup C1d2, and 23 samples from haplogroup D4h3a. 

Each of these samples is identifiable via a sample ID and the respective haplotype of said sample. In order to further ensure
anonymity, only the haplotypes were used to identify the samples in the study. Each sample for each haplogroup tends to be
unique in mutations, denoted by the Novel Variants, which are differences between the sample’s sequence and the sequenced
haplogroup. For example, mutation C3516T, which is typical for A2al, describes a transition at position 35116 in the mtDNA
genome, from a cytosine to a thymine, and 16519T, which is very common for haplogroup A2al, describes an insertion of a
thymine at position 16519.

Using the software Network, designed and made free to access by Fluxus Technology, samples and their mutations were able
to be plotted in a comprehensive phylogenetic tree, with four trees being constructed for each of the four model haplogroups.
These trees not only depict the relationship and ancestry of samples in the same haplogroup, and even samples in different
haplogroups, but can also give an estimate of time during which the haplogroup gave rise to the variants found in the tree.
For the sake of the study, we assumed an average mutation rate of one mutations every 3600 years, in accordance with
previous studies.

Table 1. An example of how each sample for each haplogroup was organized, and
their mutations read as variants.

Figure 3. A map of North and Central America depicting the
presumed migration patterns of the analyzed haplogroups,

with the divergence of A2al and B2d14 from A and B
respectively being visible on the map, signified by hollow

circles. Unlike the former two, haplogroups C1d2 and D4h3a
diverged in East Asia before these lineages reached the

Americas, hence why their migration stems from beyond
Alaska. In addition, haplogroup C1d2's approximate migration
in the FamilyTreeDNA database ends in Mexico, even though
modern-day samples are found not uncommonly in Colombia.
Thus, the rest of C1d2's migration pattern was extrapolated
based on its last known location in the database, with the

cross marking where the known migration data ends.

Discussion
This study serves as an example as to how mtDNA data can be utilized to reconstruct the history of
early peopling in different regions of the world, based on analysis of present-day inhabitants and their
mtDNA lineages, and ancient DNA isolated from remains. While we were limited to only 94 samples
total for the chosen haplogroups, further studies can and should expand upon our limited data in
order to create a more accurate and representative timeline. 

In our study, we were somewhat limited in the necessary workforce to fully derive the conclusions we
intended. While Network as a software is extremely potent in its ability to process numerous samples
and mutations, it is still an older piece of software that, as opposed to other spreadsheet software
(i.e., Microsoft Excel) requires much more manual input, and as such is significantly more time
consuming to operate. While 94 samples are hardly representative in nature, it still cost numerous
hours just to input their data into Network, thus limiting not only our time for continuing the study,
but for increasing the sample size as desired. It is also important to mention that there have likely
been numerous migrations from North America to South America across different lineages and time
periods, and that examining haplogroups of different ages allows us to better piece together the true
nature of these different waves of migrations, whereas we have only just scratched the surface.

Looking at the phylogenetic trees for each haplogroup (see Figure 2), less-centralized haplogroups
with much more variation like D4h3a tend to be older, while haplogroups that display a clear center
and less bifurcations like A2al and the others tend to be younger by comparison. This is a common
trend in other mtDNA phylogenetic trees, and corroborates the data observed in Table 2.

For the sake of the research, certain samples and mutations were excluded from the study. Samples
that were acquired from ‘Haplocall’ studies instead of other ‘Tree’ studies were excluded, for the sake
of consistency, and mutations 16519T, 64T, and 93G from haplogroup A2al, and C166111 and T152C in
haplogroup B2d14 were ignored, due to their lack of precise plotting in their respective phylogenies.

Finally, moving forward, we are hoping to compile and analyze the presence of these haplogroups in
other countries beyond Colombia, in order to create a more holistic representation for the migratory
patterns of these maternal lineages in other parts of the world.

References
Soares P, Ermini L, Thomson N, Mormina M, Rito T, Röhl A, Salas A, Oppenheimer
S, Macaulay V, Richards MB. Correcting for purifying selection: an improved
human mitochondrial molecular clock. Am J Hum Genet. 2009 Jun;84(6):740-59.
doi: 10.1016/j.ajhg.2009.05.001. Epub 2009 Jun 4. PMID: 19500773; PMCID:
PMC2694979.
Yunis JJ, Yunis EJ. Mitochondrial DNA (mtDNA) haplogroups in 1526 unrelated
individuals from 11 Departments of Colombia. Genet Mol Biol. 2013 Sep;36(3):329-
35. doi: 10.1590/S1415-47572013000300005. Epub 2013 Aug 30. PMID: 24130438;
PMCID: PMC3795164.

Luis Roniger (1997) Human Rights Violations and the Reshaping of Collective
Identities in Argentina, Chile and Uruguay, Social Identities, 3:2, 221-246, DOI:
10.1080/13504639752078 

Uricoechea Patiño, D., Collins, A., García, O. J. R., Santos Vecino, G., Cuenca, J. V.
R., Bernal, J. E., Benavides Benítez, E., Vergara Muñoz, S., & Briceño Balcázar, I.
(2023). High Mitochondrial Haplotype Diversity Found in Three Pre-Hispanic
Groups from Colombia. Genes, 14(10), 1853.
https://doi.org/10.3390/genes14101853

Ribeiro, B. P. A. (2021). Genetic characterization of the maternal lineages in
andean colombian populations (Order No. 29140543). Available from ProQuest
Dissertations & Theses Global. (2689299298). Retrieved from
https://www.proquest.com/dissertations-theses/genetic-characterization-
maternal-lineages-andean/docview/2689299298/se-2

Uricoechea Patiño, D., Collins, A., Romero García, O. J., Santos Vecino, G.,
Aristizábal Espinosa, P., Bernal Villegas, J. E., Benavides Benitez, E., Vergara
Muñoz, S., & Briceño Balcázar, I. (2023). Unraveling the Genetic Threads of
History: mtDNA HVS-I Analysis Reveals the Ancient Past of the Aburra Valley.
Genes, 14(11), 2036. https://doi.org/10.3390/genes14112036

Adriana Castillo, Fernando Rondón, Gerardo Mantilla, Leonor Gusmão, Filipa
Simão, Maternal ancestry and lineages diversity of the Santander population from
Colombia, Forensic Sciences Research, Volume 8, Issue 3, September 2023, Pages
241–248, https://doi.org/10.1093/fsr/owad032

de Saint Pierre M, Bravi CM, Motti JMB, Fuku N, Tanaka M, et al. (2012) An
Alternative Model for the Early Peopling of Southern South America Revealed by
Analyses of Three Mitochondrial DNA Haplogroups. PLOS ONE 7(9): e43486.
https://doi.org/10.1371/journal.pone.0043486

Melton, P. E. (2008). Genetic history and pre -columbian diaspora of chibchan
speaking populations: Molecular genetic evidence (Order No. 3349820). Available
from ProQuest Dissertations & Theses Global. (304617164). Retrieved from
https://www.proquest.com/dissertations-theses/genetic-history-pre-columbian-
diaspora-chibchan/docview/304617164/se-2

Díaz-Matallana, Marcela, Gómez, Alberto, Briceño, Ignacio, & Rodríguez, José
Vicente. (2016). Genetic analysis of Paleo-Colombians from Nemocón,
Cundinamarca provides insights on the early peopling of northwestern South
America. Revista de la Academia Colombiana de Ciencias Exactas, Físicas y
Naturales, 40(156), 461-483. https://doi.org/10.18257/raccefyn.328

Homburger JR, Moreno-Estrada A, Gignoux CR, Nelson D, Sanchez E, et al. (2015)
Genomic Insights into the Ancestry and Demographic History of South America.
PLOS Genetics 11(12): e1005602. https://doi.org/10.1371/journal.pgen.1005602

Table 2. Each haplogroup’s age up until the differentiation of modern variants in Colombia, with
some ages more precise than others.

Figure 2. The composite phylogenetic tree constructed by analyzing the samples of
each haplogroup, and combining each individual tree together, joined at the oldest
maternal ancestor. The center represents the hypothetical mitochondrial Eve, from
which each maternal lineage, including haplogroups A, B, C, and D, first diverged. 

https://doi.org/10.1093/fsr/owad032
https://doi.org/10.1371/journal.pone.0043486
https://doi.org/10.18257/raccefyn.328
https://doi.org/10.1371/journal.pgen.1005602