Indo-Aryan qpAdm Admixture

Table of Contents

• Introduction
• Tools Used
• Key Findings
• Methodology
• Results
  • Early Bronze Age (3300-2600 BCE): The Yamnaya Influence
  • Middle Bronze Age (2900-2350 BCE): Corded Ware
  • Late Bronze Age (1900-1200 BCE): Andronovo Culture
• Investigating BMAC Influence
• Additional Genetic Insights
  • Y-DNA Haplogroup Analysis via Big Y-700
  • IllustrativeDNA Results
• Conclusion
• Supplementary Data
  • qpAdm Model Outputs
• References

Introduction

This study presents a detailed analysis of Steppe admixture in Indo-Aryan populations, with a particular focus on the Jatt Sikh population in Punjab, across three key periods: the Early Bronze Age (3300-2600 BCE), Middle Bronze Age (2900-2350 BCE), and Late Bronze Age (2100-1200 BCE). Utilizing both personal genetic data and the Allen Ancient DNA Resource, we examine the temporal changes in genetic components, focusing on Steppe, Iranian farmer-related, and Ancient Ancestral South Indian (AASI) ancestries. Our findings reveal the stability of Steppe ancestry from the Corded Ware period onwards and the absence of direct Bactria-Margiana Archaeological Complex (BMAC) contribution, providing new insights into the genetic history of Indo-Aryan populations.

Tools Used

• ADMIXTOOLS (including qpAdm) from David Reich’s website:ADMIXTOOLS is a software package for analyzing population admixture. It includes qpAdm, a tool for estimating ancestry proportions from different source populations in a target population. This was critical for our genetic modeling. Software | David Reich Lab (harvard.edu)
• Allen Ancient DNA Resource (AADR) from David Reich’s website:AADR provides downloadable genotypes of both present-day and ancient DNA. We used version v54.1.p1 1240k for accessing necessary ancient DNA sequences. Allen Ancient DNA Resource (AADR)
• 23&me chip_v5:The 23&me chip_v5 offers over 500,000 single nucleotide polymorphisms (SNPs), which were used to obtain detailed ancestry information for the Jatt Sikh population.
• AncestryDNA data:AncestryDNA provides genetic data with over 500,000 SNPs, allowing us to cross-verify and corroborate findings from 23&me data.
• Big Y-700 for Y-DNA haplogroup confirmation:The Big Y-700 test was used to confirm Y-DNA haplogroups, specifically identifying the subclades R1a-Z93 -> R-L657 -> R-FTF40903, crucial for tracing the paternal lineage of the Jatt Sikh population.
• IllustrativeDNA G25:IllustrativeDNA G25 modeling provided additional insights and independent verification of admixture results. View the modeling results here.

Key Findings

• Primary Source of Steppe Ancestry: The main source of Steppe ancestry in South Asians is directly from the Corded Ware culture. Subsequent Steppe cultures such as Sintashta, Andronovo, and Srubnaya-Alakul did not significantly alter the existing ancestry proportions. This suggests that the genetic composition of migrating populations was largely established during the Corded Ware period. This observation aligns with findings from various studies, including those by Narasimhan et al. (2019) , which indicate that the Corded Ware culture played a pivotal role in shaping the genetic landscape of Indo-Aryan populations.
• Best Model Support: The 3-way model using Russia_Srubnaya_Alakul.SG as the Steppe source population shows the highest p-values (0.884639 for 23andMe v5 and 0.867256 for AncestryDNA), indicating it is the best fit for the data. This supports the hypothesis that Steppe admixture in Indo-Aryans predominantly originates from the Andronovo culture.
• Absence of BMAC Contribution: There is no evidence of direct genetic contribution from the Bactria-Margiana Archaeological Complex (BMAC) to the Jatt Sikh population. This aligns with other studies suggesting that the main BMAC population did not contribute significantly to South Asian gene pools. Research by Narasimhan et al. (2019) corroborates this finding, indicating that while BMAC populations mixed with Steppe communities in Central Asia, they did not significantly impact the gene pool of Indo-Aryan populations.
• Unsuccessful Models: Admixture models incorporating populations such as Turkmenistan_Gonur_BA_1, Turkmenistan_Gonur_BA_2, Uzbekistan_SappaliTepe_BA, and Turkmenistan_C_Geoksyur were not successful, as they failed to provide plausible admixture scenarios. This supports the conclusion that these populations did not significantly contribute to the genetic makeup of the Jatt Sikh population.

Methodology

For this study, we utilized genetic data from two different sources: 23andMe and AncestryDNA. By leveraging data from these two genomic companies, we aimed to enhance the robustness and confidence in our findings. The samples selected for analysis were specifically chosen to represent key periods in Indo-Aryan prehistory.

• 23andMe Data: The genetic data from 23andMe was used to obtain detailed information on the ancestry components of the Jatt Sikh population.
• AncestryDNA Data: AncestryDNA provided additional genetic data, allowing us to cross-verify and corroborate the findings from 23andMe.

Data Sources and Sample Selection

For this study, we utilized genetic data from two different sources: 23andMe and AncestryDNA. By leveraging data from these two genomic companies, we aimed to enhance the robustness and confidence in our findings. The samples selected for analysis were specifically chosen to represent key periods in Indo-Aryan prehistory.

Analytical Approach

We employed qpAdm analysis, a powerful statistical tool in population genetics, to model complex admixture scenarios. qpAdm estimates the proportions of ancestry in a target population derived from multiple source populations. This method involves two sets of populations: the “left” populations, which include the target and potential source populations, and the “right” populations, which serve as outgroups.

• Left Populations: The Jatt Sikh population and the potential source populations (e.g., Steppe, Iranian farmer-related, AASI).
• Right Populations: Outgroups used to anchor the analysis and provide a reference for comparison.

Ancestry Components and Temporal Framework

The study focused on three primary ancestral components:

1. Steppe Ancestry: Represented by various populations such as Russia_Samara_EBA_Yamnaya, Czech_CordedWare, Russia_MLBA_Sintashta, and Russia_Srubnaya_Alakul.SG.
2. Iranian Farmer-Related Ancestry: Represented by Iran_ShahrISokhta_BA2.
3. Ancient Ancestral South Indian (AASI) Ancestry: Represented by Indian_GreatAndaman_100BP.SG.

We examined the temporal changes in these genetic components across three key periods:

• Early Bronze Age (3300-2600 BCE): Russia_Samara_EBA_Yamnaya
• Middle Bronze Age (2900-2350 BCE): Czech_CordedWare
• Late Bronze Age (2100-1200 BCE): Russia_MLBA_Sintashta & Russia_Srubnaya_Alakul.SG

Statistical Validation

To ensure the accuracy and reliability of our findings, we calculated p-values for the proposed admixture models. A p-value is a statistical measure that helps determine the significance of results. In our analysis, a p-value greater than 0.05 typically indicates a plausible model, suggesting that the observed genetic patterns are statistically significant and not likely due to chance.

Results

Early Bronze Age (3300-2600 BCE): The Yamnaya Influence

The Early Bronze Age witnessed the emergence of the Yamnaya culture on the Pontic-Caspian steppe. Known for their wheeled vehicles, advanced metallurgy, and distinctive kurgan burial practices, the Yamnaya people played a crucial role in shaping Eurasian genetics and culture. Recent genomic research reveals that Yamnaya populations were composed of approximately 80% Caucasus-Lower Volga (CLV) cline and 20% Ukrainian Neolithic Hunter-Gatherers (UNHG) ancestry (Lazaridis et al., 2024). The CLV cline represents a gradient of admixture between Caucasus hunter-gatherer (CHG) ancestry and steppe populations, while the UNHG component reflects interactions with local European groups. This unique genetic profile resulted from complex interactions along three previously unrecognized Eneolithic genetic clines: the CLV cline, the Volga cline, and the Dnipro cline. The Yamnaya’s distinctive genetic makeup, combining Eastern Hunter-Gatherer (EHG) and Caucasus Hunter-Gatherer (CHG) ancestries, contributed to their cultural innovations and subsequent widespread influence across Eurasia (Haak et al., 2015).

Our qpAdm analysis for this period reveals distinct contributions from three primary ancestry components in the Indo-Aryan population:

Ancestry Component	23andMe (%)	AncestryDNA (%)
Steppe Ancestry	41.8 ± 4.4	41.8 ± 3.6
Iranian Farmer-Related Ancestry	49.0 ± 6.4	49.0 ± 5.1
Ancient Ancestral South Indian (AASI)	9.2 ± 3.0	9.2 ± 2.4

The models also include a fourth component in the 4-way analysis:

Ancestry Component	23andMe (%)	AncestryDNA (%)
Iranian Farmer-Related Ancestry	51.3% ± 6.4%	50.9% ± 4.9%
Steppe Ancestry	33.1% ± 5.9%	32.8% ± 4.6%
European Early Farmers (EEF)	6.0% ± 2.7%	6.3% ± 2.1%
Ancient Ancestral South Indian (AASI)	9.7% ± 3.0%	9.9% ± 2.2%

Key Observations:

1. Significant Steppe Ancestry: This substantial Steppe component aligns with the hypothesis of Indo-Aryan migrations into South Asia, as proposed by (Anthony, 2007). It suggests a major influx of Steppe-related populations, likely associated with the spread of Indo-European languages. The Steppe ancestry, primarily derived from Yamnaya-related groups, indicates a significant demographic impact on the ancestral Indo-Aryan population. This genetic evidence supports linguistic and archaeological theories of Indo-European expansions into South Asia during the Bronze Age.
2. Substantial Iranian Farmer-Related Component: The high proportion of Iranian farmer-related ancestry suggests long-standing genetic connections between South Asia and Iran, possibly predating Indo-European expansions (Broushaki et al., 2016). This component likely represents multiple waves of migration and interaction between Iranian plateau populations and South Asia over several millennia. It may include contributions from early agricultural communities that spread from the Fertile Crescent, as well as later Bronze Age interactions. The persistence of this component highlights the enduring influence of Near Eastern populations on South Asian genetics.
3. Presence of AASI Ancestry: The Ancient Ancestral South Indian (AASI) component underscores the resilience of indigenous South Asian genetic lineages, even in populations with significant external contributions (Basu et al., 2016). This persistence demonstrates that the Indo-Aryan migrations did not completely replace the pre-existing genetic substrate of South Asia. Instead, it suggests a process of admixture and integration between incoming groups and local populations. The AASI component represents a direct link to the earliest known inhabitants of the Indian subcontinent, preserving genetic continuity over thousands of years.
4. The presence of European Early Farmer (EEF) ancestry: This ancestry likely entered the South Asian gene pool through admixed Steppe populations, who had previously incorporated EEF ancestry in Eastern Europe. In the 4-way model suggests that Corded Ware populations, rather than direct Yamnaya descendants, were the primary source of Steppe ancestry for Indo-Aryan groups. This genetic profile reflects a multi-stage migration process involving already admixed populations from Eastern Europe, highlighting the interconnectedness of Bronze Age Eurasian societies and the complex demographic processes that shaped Indo-Aryan ancestry.

Middle Bronze Age (2900-2350 BCE): Corded Ware

During the Middle Bronze Age, significant changes occurred in the genetic landscape of Eurasia, particularly with the emergence and spread of the Corded Ware culture. This culture is believed to represent a fusion of Steppe ancestry with European Neolithic farmer populations (Haak et al., 2015). Our qpAdm analysis for this period uses Czech_CordedWare as the proxy for Steppe ancestry, providing insights into how this cultural and genetic shift affected the ancestors of the Indo-Aryan population.

Before examining the Indo-Aryan ancestry, we first analyzed the Czech_CordedWare population itself:

Ancestry Component	Percentage (%)
Russia_Samara_EBA_Yamnaya	71.1% ± 1.5%
Ukraine_EBA_GlobularAmphora	28.9% ± 1.5%

This model suggests that the Czech Corded Ware population was primarily derived from Yamnaya-related steppe ancestry, with a significant contribution from earlier European farming populations (Allentoft et al., 2015).

Using Czech_CordedWare as a proxy for Steppe ancestry, our 3-way qpAdm model for the Jatt Sikh population during the Middle Bronze Age yields:

Ancestry Component	23andMe (%)	AncestryDNA (%)
Iranian Farmer-Related Ancestry	56.2 ± 5.4	56.3 ± 4.3
Steppe Ancestry	35.3 ± 3.5	34.9 ± 2.8
Ancient Ancestral South Indian (AASI)	8.5 ± 2.9	8.8 ± 2.3

Key observations:

• Increase in Iranian-related ancestry: The Iran_ShahrISokhta_BA2 component shows a significant increase from the Early Bronze Age model, now constituting the majority of the ancestry at about 56%. This increase is likely due to the incorporation of Anatolian farmer-related ancestry from the European Early Farmers (EEF) component in the Corded Ware population.
• Hypothesis on EEF and Iranian-related ancestry: The increase in Iran_ShahrISokhta_BA2 ancestry can be explained by the high Anatolian farmer ancestry in EEF populations. When Steppe populations (like Yamnaya) admixed with EEF groups to form the Corded Ware culture, they incorporated this Anatolian-related ancestry (Haak et al., 2015). In our qpAdm models, this additional Anatolian-related ancestry is likely being captured by the Iran_ShahrISokhta_BA2 component, leading to its apparent increase.
• Steppe ancestry: The Czech_CordedWare component, representing Steppe ancestry, shows a slight decrease compared to the Yamnaya-related ancestry in the Early Bronze Age model, now at about 35%. This reflects the admixed nature of the Corded Ware population itself, which includes both Steppe and European farmer ancestries (Allentoft et al., 2015).
• AASI component: The Ancient Ancestral South Indian (AASI) component, represented by Indian_GreatAndaman_100BP.SG, shows a slight decrease to about 8-9%, but remains a significant part of the genetic makeup (Narasimhan et al., 2019).
• Model consistency: Both 23andMe and AncestryDNA datasets show remarkably consistent results, increasing confidence in the model’s reliability.

These findings align with current understanding of Bronze Age population dynamics in Eurasia and highlight the complex nature of genetic admixture during this period. The significant increase in Iranian-related ancestry, likely due to the incorporation of Anatolian farmer-related ancestry via the EEF component in Corded Ware populations, demonstrates the intricate genetic history of the ancestors of the Indo-Aryan population.

Late Bronze Age (1900-1200 BCE): Andronovo Culture

For the Late Bronze Age period, we used Russia_Srubnaya_Alakul.SG as the proxy for Steppe ancestry. This population represents a later phase of the Steppe-related groups that likely contributed to Indo-Aryan gene pools, aligning with the Western Steppe Middle to Late Bronze Age (Steppe MLBA) ancestry cluster (Narasimhan et al., 2019).

Ancestry Component	23andMe (%)	AncestryDNA (%)
Iranian Farmer-Related Ancestry	58.3 ± 5.1	59.1 ± 4.1
Steppe Ancestry	35.8 ± 3.5	34.7 ± 2.8
Ancient Ancestral South Indian (AASI)	5.9 ± 2.8	6.2 ± 2.2

Key Observations:

• Model Fit: Both models show excellent statistical fit, with very high p-values (0.884639 for 23andMe and 0.867256 for AncestryDNA). This indicates that the model is highly plausible for explaining the ancestry of the Jatt Sikh population.
• Iranian-related Ancestry: The Iran_ShahrISokhta_BA2 component remains the largest contributor, accounting for about 58-59% of the ancestry. This is consistent with findings from (Narasimhan et al., 2019), which emphasized the importance of Iranian farmer-related ancestry in South Asian populations. This component likely represents a complex mixture of earlier Iranian-related ancestries and Anatolian farmer-related ancestry incorporated through admixture events in the Steppe (Haak et al., 2015).
• Steppe Ancestry: The Russia_Srubnaya_Alakul.SG component contributes approximately 35-36% of the ancestry. This substantial Steppe-related contribution aligns with the hypothesis of significant Indo-Aryan migrations during this period, as proposed by (Anthony, 2007) and supported by genetic evidence in (Narasimhan et al., 2019). The Srubnaya culture, part of the Steppe MLBA cluster, represents a population with a mixture of Yamnaya-related ancestry and European farmer ancestry (Wang et al., 2019).
• AASI Component: The Ancient Ancestral South Indian (AASI) ancestry, represented by Indian_GreatAndaman_100BP.SG, accounts for about 6% of the total ancestry. This persistent AASI component, albeit at a lower proportion than in earlier periods, is consistent with the findings of (Shinde et al., 2019) regarding the continuity of indigenous South Asian genetic lineages.
• Temporal Context: The use of Russia_Srubnaya_Alakul.SG (1900-1200 BCE) as the Steppe proxy aligns well with the estimated timeframe for significant Steppe ancestry entering South Asia (Narasimhan et al., 2019). This period coincides with the proposed timing of Indo-European language spread into South Asia, supporting linguistic and archaeological theories of Indo-Aryan expansions.

These results provide a nuanced picture of the genetic history of the Indo-Aryan population during the Late Bronze Age. The significant Steppe ancestry component, coupled with the predominant Iranian-related ancestry and a persistent AASI component, reflects the complex, multi-layered nature of Indo-Aryan genetic heritage. This genetic profile is consistent with the broader pattern observed across South Asia, where Steppe-related ancestry was integrated into pre-existing population structures characterized by a mixture of Iranian farmer-related and indigenous South Asian ancestries (Narasimhan et al., 2019).

Investigating BMAC Influence

To further investigate potential genetic contributions from Central Asia and to test the hypothesis of BMAC influence on Indo-Aryan populations, we attempted to incorporate Bactria-Margiana Archaeological Complex (BMAC) populations into our qpAdm models. The BMAC, also known as the Oxus civilization, was a significant Bronze Age culture in Central Asia that had known interactions with both Steppe and South Asian populations.

BMAC sources:

• Turkmenistan_Gonur_BA_1: Bronze Age sample from Gonur Tepe, Turkmenistan (2300-1600 BCE), associated with the BMAC.
• Turkmenistan_Gonur_BA_2: Another Gonur Tepe sample (2300-1600 BCE), representing a major BMAC urban center.
• Uzbekistan_SappaliTepe_BA: Bronze Age sample from southern Uzbekistan (2100-1700 BCE), marking BMAC’s northernmost extent.
• Turkmenistan_C_Geoksyur: Chalcolithic sample from southeastern Turkmenistan (3200-2800 BCE), predating and influencing BMAC.

All models incorporating BMAC populations proved infeasible due to either negative coefficients or excessively large standard errors. This outcome suggests that there was no significant direct BMAC contribution to Jatt Sikh ancestry.

The failure of BMAC-inclusive models aligns with findings by (Narasimhan et al., 2019), which indicated that BMAC populations did not substantially impact South Asian gene pools. This result supports the hypothesis that while BMAC populations interacted with both Steppe and South Asian groups, they did not contribute significantly to the genetic makeup of Indo-Aryan populations.

Additional Genetic Insights

While our qpAdm analysis provides a comprehensive view of the genetic admixture in the Indo-Aryan population across different Bronze Age periods, additional genetic data can offer further insights and corroboration. To this end, we examined personal Y-DNA data and conducted G25 IllustrativeDNA modeling, which provide independent lines of evidence to support our findings.

Y-DNA Haplogroup Analysis via Big Y-700

The Big Y-700 test revealed a Y-DNA haplogroup of R-FTF40903, a subclade of R1a-Z93. This haplogroup, particularly its R-L657 subclade, is strongly associated with Indo-Aryan migrations into South Asia. The chronological story of this lineage unfolds as follows:

1. Origin in the Steppe: R1a likely originated in the Pontic-Caspian steppe around 5000 BCE.
2. Corded Ware Expansion: The Z93 subclade emerged circa 3000 BCE, coinciding with the Corded Ware culture’s spread.
3. Indo-Aryan Migrations: R-L657, a key marker of Indo-Aryan expansions, likely arose around 2500 BCE.
4. South Asian Presence: By 1500 BCE, R-L657 had become prevalent in South Asia, aligning with the proposed timing of Indo-Aryan arrivals.

This Y-DNA evidence supports our qpAdm results showing significant Steppe ancestry in the Jatt Sikh population.

IllustrativeDNA Results

IllustrativeDNA’s two-way admixture model suggests:

1. Andronovo Culture: 34.1%
2. Indus Valley Civilization: 65.9%

This model closely aligns with our qpAdm analysis, particularly the Late Bronze Age results:

• The Andronovo component (34.1%) corresponds well with the Steppe ancestry in our qpAdm model (35.8% in 23andMe and 34.7% in AncestryDNA).
• The Indus Valley Civilization component (65.9%) likely represents a combination of the Iranian farmer-related ancestry and AASI component from our qpAdm models.

These independent analyses provide additional support for our qpAdm findings, reinforcing the reliability of our conclusions regarding Jatt Sikh ancestry and its connection to Indo-Aryan migrations.

Conclusion

Our qpAdm analysis of Indo-Aryan ancestry across the Bronze Age periods provides significant insights into the complex genetic history of Indo-Aryan populations in South Asia. The results support a nuanced understanding of Steppe migrations and their impact on the genetic landscape of the region. Key findings:

1. Steppe Ancestry Stability: The Steppe ancestry component remains relatively constant at approximately 35% from the Middle Bronze Age (2900-2350 BCE) through the Late Bronze Age (1900-1200 BCE). This stability suggests that the genetic composition of the migrating populations was largely established during the Corded Ware period, with minimal changes in subsequent Steppe cultures (Sintashta, Andronovo, Srubnaya-Alakul).
2. Andronovo as Primary Vector: The 3-way model using Russia_Srubnaya_Alakul.SG as the Steppe source population shows the highest p-values, supporting the hypothesis that the Andronovo culture was likely the immediate vector for Steppe migration into South Asia. However, this population already carried the genetic signature established during the Corded Ware period.
3. Early Admixture Events: The most significant change in Steppe ancestry proportion occurs between the Early and Middle Bronze Age, decreasing from ~42% to ~35%. This change coincides with the expansion of Yamnaya-derived populations and their admixture with European farmers, forming the Corded Ware culture.
4. Increase in Iranian-related Ancestry: The Iran_ShahrISokhta_BA2-related ancestry increases from ~49% to ~56% between the Early and Middle Bronze Age. This increase likely results from the incorporation of Anatolian farmer-related ancestry via European Early Farmers in the Corded Ware population.
5. Absence of Direct BMAC Contribution: Our qpAdm models show no evidence of direct genetic contribution from the Bactria-Margiana Archaeological Complex (BMAC) to the Jatt Sikh population. This aligns with recent studies suggesting minimal BMAC impact on South Asian gene pools.
6. Persistence of Indigenous Ancestry: The Ancient Ancestral South Indian (AASI) component, while decreasing slightly over time, remains a significant part of the genetic makeup, demonstrating the resilience of indigenous South Asian lineages.

These findings paint a picture of Indo-Aryan ancestry as the product of complex, multi-stage migration and admixture processes. The genetic profile was largely shaped by Steppe migrations, particularly those associated with the Corded Ware and subsequent Andronovo cultures. However, the significant Iranian-related ancestry and persistent AASI component highlight the importance of pre-existing South Asian genetic substrates and earlier Near Eastern influences.

Our study contributes to the growing body of evidence supporting a nuanced view of Indo-Aryan migrations, emphasizing the importance of earlier admixture events in shaping the genetic landscape of South Asia. It underscores the complex interplay between Steppe, Near Eastern, and indigenous South Asian populations in forming the genetic tapestry of modern Indo-Aryan groups.

Supplementary Data

qpAdm Model Outputs:

1. sohi_23andme_Russia_Samara_EBA_Yamnaya
2. sohi_ancestryDNA_Russia_Samara_EBA_Yamnaya
3. sohi_23andme_Czech_CordedWare
4. sohi_ancestryDNA_Czech_CordedWare
5. sohi_23andme_Russia_Srubnaya_Alakul
6. sohi_ancestryDNA_Russia_Srubnaya_Alakul
7. sohi_23andmeV5_3_way_qpadm_Sintashta
8. sohi_ancestry_3_way_qpadm_Sintashta
9. sohi_23andme_Russia_MBA_Poltavka
10. sohi_23andme_Turkmenistan_Gonur_BA_1
11. sohi-23andme_Mongolia_EIA_SlabGrave_1
12. sohi_23andme_Kazakhstan_Kumsay_EBA
13. sohi_23andme_BMAC
14. sohi_ancestryDNA_BMAC
15. sohi_23andmeV5_4way_qpadm_Gonur_BA1
16. sohi_ancestry_4way_qpadm_Gonur_BA1
17. sohi_23andme_4way_qpadm_Gonur_BA2
18. sohi_ancestry_4way_qpadm_Gonur_BA2
19. sohi_23andme_4way_qpadm_Geoksyur
20. sohi_ancestry_4way_qpadm_Geoksyur

References

1. Allen Ancient DNA Resource (AADR)
2. Haak, Wolfgang, et al. “Massive migration from the steppe was a source for Indo-European languages in Europe.” Nature 522.7555 (2015): 207. Link to Article
3. Narasimhan, Vagheesh M., et al. “The formation of human populations in South and Central Asia.” Science 365.6457 (2019). Link to Article
4. Lazaridis, Iosif, et al. “Genomic insights into the origin of farming in the ancient Near East.” Nature 536.7617 (2024): 419-424. Link to Article
5. Anthony, David W. The Horse, the Wheel, and Language: How Bronze-Age Riders from the Eurasian Steppes Shaped the Modern World. Princeton University Press, 2007. Link to Book
6. Broushaki, Farnaz, et al. “Early Neolithic genomes from the eastern Fertile Crescent.” Science 353.6298 (2016): 499-503. Link to Article
7. Basu, Analabha, et al. “Genomic reconstruction of the history of extant populations of India reveals five distinct ancestral components and a complex structure.” Proceedings of the National Academy of Sciences 113.6 (2016): 1594-1599. Link to Article
8. Allentoft, Morten E., et al. “Population genomics of Bronze Age Eurasia.” Nature 522.7555 (2015): 167-172. Link to Article
9. Wang, Chuan-Chao, et al. “The genetic prehistory of the Greater Caucasus.” Nature Communications 10.1 (2019): 1-12. Link to Article
10. Shinde, Vasant, et al. “An ancient Harappan genome lacks ancestry from Steppe pastoralists or Iranian farmers.” Cell 179.3 (2019): 729-735. Link to Article