Brahmin DNA and Genetics: What the Science Actually Says

Few words in Indian life carry as much weight, or as much argument, as "Brahmin". It names a varna in ancient texts, hundreds of living communities from Srinagar to Thiruvananthapuram, and, more and more often, a label in genetics papers. Online, it also attracts confident slogans: Brahmins are "Aryans", Brahmins are "pure".

The peer-reviewed studies are more interesting than any of the slogans. They show a real, measurable pattern in how steppe-related ancestry is spread across India. They also show that Brahmin communities are not one population but many, shaped by their regions and by centuries of marrying within their own jati. This guide walks through what the studies actually found, quoting them carefully, and is just as clear about what DNA cannot do.

Varna, jati and gotra: what "Brahmin" actually refers to

Before any genetics, it helps to separate three words that are often blurred together. Genetics can only see one of them clearly.

Varna is a classification in texts

Varna is the fourfold social classification described in Brahmanical literature: Brahmins (priests, scholars and teachers), Kshatriyas, Vaishyas and Shudras. It is usually traced to the Purusha Sukta hymn of the Rigveda and is set out in texts such as the Manusmriti. Scholars describe it as a theoretical or idealised scheme, and many Hindu texts and traditions themselves questioned it.

Jati is the community people actually marry within

Jati refers to the thousands of endogamous communities that exist on the ground. "Brahmin" is a varna label that covers a great many jatis: Iyer, Iyengar, Namboothiri, Chitpavan, Deshastha, Saraswat, Maithil and many more, each with its own language, region, customs and marriage networks.

This matters for DNA. Geneticists define a population by who has children with whom, which is a jati question, not a varna question. When a paper says "Brahmins", it means specific sampled jatis from particular places.

Gotra is a lineage name, not a population

A gotra is a lineage traced through the male line to a common ancestor, traditionally one of the rishis. Custom forbids marriage within the same gotra, and the same gotra names are found across different castes. In principle a paternal lineage could leave a trace on the Y chromosome, but we are not aware of any peer-reviewed study showing that gotras correspond to distinct Y-DNA lineages. Our guide to gotra and DNA looks at this in more detail.

Pancha Gauda and Pancha Dravida

Brahmins have long recognised regional groupings themselves. Kalhana's Rajatarangini, a 12th-century chronicle of Kashmir, divides Brahmins into the Pancha Gauda north of the Vindhyas (Sarasvata, Kanyakubja, Gauda, Utkala and Maithila) and the Pancha Dravida to the south (Karnataka, Tailanga, Dravida, Maharashtraka and Gurjara). "Brahmin" has always meant a family of distinct regional communities.

The ancestry framework: ANI, ASI, AASI and steppe

Every finding about Brahmin genetics sits inside a bigger picture of how all Indians came to be who they are.

Two ancestral populations

In 2009, Reich and colleagues published a landmark study in Nature that genotyped 132 people from 25 Indian groups. They found that most Indians descend from a mixture of two ancient populations. The "Ancestral North Indians" (ANI) were genetically close to Middle Easterners, Central Asians and Europeans. The "Ancestral South Indians" (ASI) were as distinct from ANI and East Asians as those are from each other.

ANI ancestry ranged from 39% to 71% in most Indian groups, and the study found it "is higher in traditionally upper caste and Indo-European speakers". It also noted that groups with only ASI ancestry may no longer exist in mainland India. In other words, almost everyone in India is a mixture, and the differences between communities are differences of proportion. For a fuller explanation, see our guide to steppe ancestry in India.

Ancient DNA sharpened the picture

A decade later, Narasimhan and colleagues published the largest ancient DNA study of the region in Science (2019), with genome-wide data from 523 ancient individuals analysed alongside 1,789 present-day people from 246 South Asian groups. They identified three deep sources behind modern South Asian ancestry:

• AASI (Ancient Ancestral South Indian), a lineage of the subcontinent's early hunter-gatherers, deeply separated from other Eurasian lineages.
• Iranian farmer-related ancestry, carried by people on what the authors call the Indus Periphery Cline, who were 45-82% Iranian farmer-related and 11-50% AASI.
• Steppe ancestry from Middle to Late Bronze Age pastoralists of the Central Asian steppe, which the study says "integrated further south in the first half of the second millennium BCE, contributing up to 30% of the ancestry of modern groups in South Asia."

Crucially, ANI and ASI were themselves mixtures that formed after about 2000 BCE. The authors estimate a minimum of about 70% Indus Periphery Cline ancestry for the ANI and about 55% for the ASI. Since the Indus Periphery people already carried AASI ancestry, so did the ANI. A separate 2019 study in Cell by Shinde and colleagues sequenced a genome from Rakhigarhi, an Indus Valley Civilization site, and found it had little if any steppe pastoralist ancestry, showing that this ancestry was not widespread in northwest South Asia during the Indus period as it is today.

Brahmin groups are not outside this framework. They sit on the same gradient as everyone else, generally toward the end with more ANI, and exactly where depends heavily on region.

The 2019 steppe finding, stated precisely

The single most quoted result in this field is a section of the Narasimhan paper titled "Steppe Ancestry in South Asia is Primarily from Males and Disproportionately High in Brahmins". Because it is often exaggerated, here is exactly what it says.

What the authors did and found

The team took 140 present-day South Asian groups that fit their main model and asked whether any had more steppe-related ancestry, relative to Indus Periphery-related ancestry, than the model predicts. Six groups stood out at a very strict statistical threshold. The two strongest signals were in a group labelled Brahmin_Tiwari and in Bhumihar_Bihar.

In the authors' words, five of those six groups "were Brahmins or Bhumihars even though they comprise only 7-11% of the 140 groups analyzed." Their overall summary is that groups who view themselves as being of traditionally priestly status "tend (with exceptions) to have more Steppe ancestry than expected based on ANI-ASI mixture".

They read this as an independent line of evidence, alongside the link between Bronze Age Eastern Europe and South Asia, that South Asia's Indo-European languages arrived with people from the steppe, since Brahmins were the traditional custodians of liturgy in Sanskrit.

The caveats the paper itself gives

• "Not a formal test." Groups share ancestry, so they are not statistically independent, and the choice of groups was not made blind (four "Catholic Brahmin" groups were excluded).
• "With exceptions." Not every Brahmin group stood out. It is a tendency, not a rule.
• The signal is not confined to Brahmins. One of the six outlier groups was neither Brahmin nor Bhumihar, and steppe ancestry contributes to many South Asian groups of every background.
• It is an excess over a model, not a majority. With steppe ancestry contributing up to about 30% in modern groups, even the groups with the strongest signal remain mostly a mix of Iranian farmer-related and AASI-related ancestry.
• It is about groups, not people. It compares averages of sampled populations, not individuals.

The same section adds a detail about how steppe ancestry arrived. In present-day South Asians it is over-represented on the Y chromosome compared with the rest of the genome, so its introduction "was mediated mostly by males".

So, are Brahmins "Aryans"?

"Arya" was a self-designation used by Indo-Iranian-speaking peoples in the Rigveda and the Avesta. Many modern scholars see that identity as religious, cultural and linguistic, not racial. The idea of an "Aryan race" is a 19th-century European invention, popularised by Arthur de Gobineau and later used by racial supremacists.

Genetics can measure steppe-related ancestry, and the 2019 study links that ancestry to the spread of Indo-European languages. It cannot measure who called themselves arya three thousand years ago. What the data does show is that every Brahmin group studied is a mixture, that the steppe share is a minority even where it is highest, and that many non-Brahmin groups carry steppe ancestry too. "Aryan" is a word about language and culture, not a genetic category. Our article on the Aryan migration debate and DNA covers the wider history.

Brahmin haplogroups: R1a-Z93 and the rest

Haplogroups trace a single line of descent: Y-DNA from father to son, mitochondrial DNA (mtDNA) from mother to child. Each is one ancestral line among thousands.

R1a on the paternal side

The Y-DNA haplogroup most associated with Brahmins is R1a, and specifically the subgroup R1a-Z93. A 2009 study in the Journal of Human Genetics by Sharma and colleagues compiled 2,809 Y chromosomes, including 681 from Brahmins, and reported the highest frequency of R1a1* "(up to 72.22%)" in Brahmins. The authors argued for an Indian origin of this lineage.

Ancient DNA has since added important context. The 2019 Science study found R1a-Z93 at 100% frequency among the Middle to Late Bronze Age steppe males it analysed, and found that steppe ancestry entered present-day South Asian groups mainly through men. That fits R1a-Z93 having arrived largely with the steppe movement. Our guide to the R1a haplogroup in India covers the debate in more depth.

Earlier studies agree at a broader level. Bamshad and colleagues (Genome Research, 2001) found Y-DNA affinity to Europeans rising with caste rank in Andhra Pradesh, and Watkins and colleagues (BMC Genetics, 2008) found Tamil Brahmins had high Y-chromosome affinity to eastern Europeans through R1a1.

But never only R1a

The "up to 72%" figure is the highest value in one compilation, not a typical Brahmin frequency. Other samples look quite different:

• A 2026 study by Dave and colleagues of 421 Gujarati Brahmin men predicted haplogroups from Y-STR markers and found R1a at 39%, followed by L at 15% and J at 12%.
• A 2021 study by Mahal of Brahmin Y-DNA concluded that the paternal lineages in its dataset trace to at least 12 different regions, with about 83% belonging to four major haplogroups: two that emerged from Central Asia, one from the Fertile Crescent and one of indigenous Indian origin.
• R1a is not a Brahmin marker either. Kivisild and colleagues (American Journal of Human Genetics, 2003) found R1a at its highest frequency in Punjab and at 26% in the Chenchu, a tribal community of south India.

So an R1a result is one interesting fact about one paternal line. It is not evidence of caste, and plenty of Brahmin men carry other haplogroups entirely.

Maternal lines tell a different story

On the maternal side the pattern changes. Bamshad's 2001 study found that for mtDNA each caste was most similar to Asians, with 20-30% of Indian mtDNA haplotypes belonging to West Eurasian haplogroups, the highest share in upper castes. Watkins in 2008 likewise found that Tamil castes, Brahmins included, had higher mtDNA affinity to eastern Asians than to Europeans. Kivisild's 2003 study showed that Indian-specific branches of mtDNA haplogroups M and N are shared by populations of every social rank across the subcontinent.

Brahmin maternal lines are mostly rooted in the subcontinent, like those of their neighbours: a story of mixing, not separation.

Endogamy: why Brahmin communities stayed genetically distinct

If Brahmins are a mixture like everyone else, why can geneticists tell communities apart at all? The answer is endogamy, the custom of marrying within one's own community.

• Reich et al. (2009) found that allele frequency differences between groups in India are larger than in Europe, reflecting "strong founder effects whose signatures have been maintained for thousands of years owing to endogamy".
• Moorjani et al. (American Journal of Human Genetics, 2013) analysed 73 groups and dated ANI-ASI mixing to about 1,900-4,200 years ago. After that, India shifted from "a region in which major population mixture was common to one in which mixture even between closely related groups became rare because of a shift to endogamy".
• Basu et al. (PNAS, 2016) studied 367 people from 20 populations and found that widespread mixing "was rapidly replaced by endogamy about 70 generations ago, among upper castes and Indo-European speakers predominantly", a time they note coincides with the formulation of social norms restricting intermarriage.
• Kerdoncuff et al. (Cell, 2025) sequenced the whole genomes of 2,762 people across India and found extensive homozygosity and shared DNA segments reflecting strong founder events from a recent shift towards endogamy.

Each jati that married mostly within itself for centuries developed its own small genetic signature through drift, and two Brahmin communities a thousand kilometres apart drift in different directions. That is a large part of why "Brahmin" is not one genetic population. Our article on how caste endogamy shaped Indian DNA explores this across all communities.

One word, many populations: Brahmin diversity across India

The clearest lesson from the research is that "Brahmin" is a social category stretched across very different regional populations. Three findings make the point:

• In Reich's 2009 data, the Kashmiri Pandit sample sat at the far end of the gradient, the group most related to Europeans among those studied. Other upper caste groups sat further along the cline.
• The 2019 steppe pattern holds "with exceptions", so Brahmin groups do not all share the same steppe share.
• Watkins and colleagues (2008) sampled Tamil-speaking Brahmins in and near Chennai, and a second Brahmin group in Kanchipuram that included Kannada- and Telugu-speaking families whose ancestors had moved from Karnataka and Andhra Pradesh. The linguistically different group "showed several differences" from the Tamil castes in their analysis. Two Brahmin samples from the same corner of Tamil Nadu were not genetically interchangeable.

Are Brahmins closer to their neighbours than to each other?

This is often claimed. We looked for a peer-reviewed, genome-wide study that directly tests whether Brahmins of one region are closer to their non-Brahmin neighbours than to Brahmins elsewhere, and did not find one that settles the question. Some data even points the other way: Watkins found that genetic variation between castes within Tamil Nadu or Andhra Pradesh was larger than the variation between comparable groups from the two states.

What is well supported is narrower. Brahmin groups share deep ancestry with their regional neighbours, since all of them carry AASI and Iranian farmer-related ancestry. Within Tamil Nadu, Watkins found genetic differentiation among castes was low, "reflecting a largely common origin". And Brahmin groups from different regions differ from one another in their proportions of each ancestry.

North and northwest India

This region includes the Kashmiri Pandits, the Saraswat Brahmins of Punjab, the Mohyal Brahmins, and the large Brahmin communities of Uttar Pradesh and Rajasthan, including groups such as the Gaur Brahmins named in the old Pancha Gauda list. Read more in our articles on Kashmiri Pandit DNA and ancestry and on the Sharma surname, gotra and DNA.

East India

Maithil Brahmins of Bihar, the Brahmin communities of Bengal, Odisha and Assam, and the Nepali Bahun community all belong here. A Bhumihar group from Bihar gave one of the two strongest signals in the 2019 steppe analysis.

West India and the Konkan

The Chitpavan (Kokanastha) and Deshastha Brahmins of Maharashtra, Konkani Brahmins and the Goud Saraswat Brahmins of the coast form a distinct western cluster of communities. Genetics touches them indirectly: a 2021 study in Human Genetics by Kumar and colleagues analysed 110 Roman Catholics from Goa, Kumta and Mangalore, a community some historians have linked to Gaud Saraswat Brahmins, and found close affinity with Indo-European-speaking groups of India, "particularly Brahmins". It is a striking reminder that genetic similarity and present-day identity are different things.

South India

The Iyers and Iyengars of Tamil Nadu, the Namboothiris of Kerala, and the Telugu and Kannada Brahmin communities are classed as Pancha Dravida in the old scheme. For a closer look at two Tamil communities, see Iyer vs Iyengar DNA.

Brahmin reference groups in our panel

Helixline compares your DNA with more than 140 South Asian community reference groups. These are the Brahmin groups among them:

Region	Brahmin reference groups in the Helixline panel
North and northwest	Kashmiri Pandit, Saraswat Brahmin (Punjab), Mohyal Brahmin, Uttar Pradesh Brahmin, Rajasthani Brahmin
East and northeast	Maithil Brahmin, Bihar Brahmin, Bengali Brahmin, Odia Brahmin, Assamese Brahmin, Nepali Brahmin (Bahun)
West and Konkan	Chitpavan/Kokanastha Brahmin, Deshastha Brahmin, Konkani Brahmin, Goud Saraswat Brahmin
South	Tamil Brahmin (Iyer), Tamil Brahmin (Iyengar), Namboothiri Brahmin, Telugu Brahmin, Kannada Brahmin

The panel also includes the Bhumihar community. Each of these is a reference group for similarity comparisons, not a category anyone is placed into.

What the research adds up to, and what DNA cannot do

The studies at a glance

Study	Year	Sample	Finding relevant to Brahmin communities
Bamshad et al., Genome Research	2001	About 265 men, eight castes	Y-DNA affinity to Europeans rises with caste rank; mtDNA closest to Asians
Kivisild et al., AJHG	2003	Two tribes, six caste groups	Indian-specific mtDNA shared across ranks; R1a 26% in the Chenchu tribe
Watkins et al., BMC Genetics	2008	155 people, four Tamil castes	Low differentiation among Tamil castes; Kannada- and Telugu-speaking Brahmins differed
Sharma et al., Journal of Human Genetics	2009	2,809 Y chromosomes, 681 from Brahmins	R1a1* up to 72.22% in Brahmins
Reich et al., Nature	2009	132 people from 25 groups	ANI 39-71%, higher in upper castes and Indo-European speakers
Moorjani et al., AJHG	2013	73 groups	ANI-ASI mixing 1,900-4,200 years ago, then a shift to endogamy
Basu et al., PNAS	2016	367 people from 20 populations	Endogamy replaced mixing about 70 generations ago, mainly among upper castes and Indo-European speakers
Narasimhan et al., Science	2019	523 ancient individuals; 246 present-day groups	5 of 6 groups with excess steppe ancestry were Brahmin or Bhumihar, "with exceptions"
Shinde et al., Cell	2019	One genome from Rakhigarhi	Little if any steppe ancestry in an Indus Valley Civilization individual
Mahal, Molecular Genetics and Genomics	2021	Brahmin Y-DNA dataset	Paternal lineages from at least 12 regions
Kerdoncuff et al., Cell	2025	2,762 whole genomes	Three main ancestral sources; strong founder events from endogamy
Dave et al., International Journal of Legal Medicine	2026	421 Gujarati Brahmin men	Predicted Y haplogroups: R1a 39%, L 15%, J 12%

DNA cannot identify or prove a caste

Caste is a social identity. Genes do not carry a jati label, and no test reads one. What a test can do is compare your DNA with reference groups and report how similar you are to each. Those results are similarity comparisons, not membership.

Neighbouring communities share most of their deep ancestry, families have changed identity and region over centuries, and, as the Goan Catholic study shows, a community can be genetically closest to Brahmin groups while having an entirely different identity today. No test can settle a family question about caste.

DNA cannot rank anyone

A share of steppe, Iranian farmer-related or AASI ancestry is a record of where some of your ancestors lived thousands of years ago. It says nothing about worth, ability or status.

"Purity" is not supported

The idea that any Brahmin community is genetically "pure" does not survive contact with the data. Brahmin groups sit on the same ANI-ASI gradient as other Indians. The ANI were themselves at least about 70% Indus Periphery-related, and those people carried 11-50% AASI ancestry. Even the steppe arrivals were a mixture: the 2019 study describes them as descending from mixtures of people related to Yamnaya steppe pastoralists and European farmers, who then "mixed further with local populations" in South Asia to form the ANI. When every source population was itself a mix, "purity" has no meaning.

Testing as a Brahmin family: what a Helixline report shows

A personal test shows your own proportions, not a group average.

For families in India

Helixline kits use a saliva sample, ship within India as standard and include free doorstep pickup of your sample. The kit reaches you in 3-4 days, and results are ready in 2-3 weeks from when we receive your sample at our lab, which is an ISO 15189 certified Indian lab partner. Your data stays in India.

Kit	Price	What it includes
Origins	₹6,999	State-level ancestry breakdown, community similarity, ANI/ASI/steppe/AASI composition, Y-DNA and mtDNA haplogroups
Decode	₹12,999	Everything in Origins plus health, traits and pharmacogenomics reports and a Health Coach consultation
Infinite	₹17,999	Everything in Decode plus a 137-marker Full Body Blood Test and a doctor consultation

Your ancestry results are built on 7,600+ global reference samples and 140+ South Asian community reference groups, including the 20 Brahmin groups listed above. Y-DNA results need a male tester, so women can learn their paternal line through a father or brother. Order the Decode kit or compare all kits on the buy page.

For Brahmin families abroad

If you already tested with 23andMe, AncestryDNA, MyHeritage or FamilyTreeDNA, you do not need a new kit. Upload your raw data for Upload Ancestry at $25 / £19 / €23 / CA$34 / A$38 / AED 92 / S$34, or Upload Complete at $50 / £39 / €46 / CA$68 / A$76 / AED 184 / S$68. Results arrive within 1-2 days. Kits ship within India as standard; if you would rather test with a new kit where you live, delivery abroad can be arranged on request with a quote through our international page. You can also gift a kit to parents in India by choosing "Someone in India" at checkout on the buy page, or see our gifting page.

Sources and further reading

• Reich et al. (2009), Nature
• Narasimhan et al. (2019), Science
• Shinde et al. (2019), Cell
• Moorjani et al. (2013), American Journal of Human Genetics
• Basu et al. (2016), PNAS
• Kerdoncuff et al. (2025), Cell
• Bamshad et al. (2001), Genome Research
• Kivisild et al. (2003), American Journal of Human Genetics
• Watkins et al. (2008), BMC Genetics
• Sharma et al. (2009), Journal of Human Genetics
• Mahal (2021), Molecular Genetics and Genomics
• Dave et al. (2026), International Journal of Legal Medicine
• Kumar et al. (2021), Human Genetics
• Background (Wikipedia): Varna, Gotra, Pancha Gauda, Pancha Dravida, Aryan

Frequently Asked Questions

Do Brahmins have more steppe ancestry than other Indians?

On average, some do. A 2019 study in Science by Narasimhan and colleagues found six South Asian groups with more steppe-related ancestry than a simple ANI-ASI model predicts, and five of the six were Brahmin or Bhumihar groups. The authors stress that this holds with exceptions and is not a formal test. Steppe ancestry contributes up to about 30% in many South Asian groups, not only Brahmins, and is a minority share even where it is highest.

What is the most common Brahmin haplogroup?

On the paternal side, R1a, mostly R1a-Z93, is frequent in many Brahmin samples, with one 2009 compilation reporting up to 72.22%. Frequencies vary widely: a 2026 study of 421 Gujarati Brahmin men found R1a at 39%, L at 15% and J at 12%. Brahmin maternal lineages are mostly Indian-specific mtDNA branches shared with neighbouring communities.

Are Brahmins Aryans according to DNA?

Genetics cannot answer that in the form it is usually asked. Arya was a cultural, religious and linguistic self-designation in the Rigveda and the Avesta, not a genetic category. DNA can measure steppe-related ancestry, which some Brahmin groups carry more of than predicted, but every Brahmin group studied is a mixture of steppe, Iranian farmer-related and AASI ancestry, and many non-Brahmin groups carry steppe ancestry too.

Are all Brahmins genetically the same?

No. Brahmin is a social category covering many regional jatis, each of which has largely married within itself for centuries. Reich and colleagues found the Kashmiri Pandit sample at the far end of India's ancestry gradient, the 2019 steppe pattern holds only with exceptions, and a 2008 Tamil Nadu study found a Kannada- and Telugu-speaking Brahmin sample differed from Tamil castes.

Can a DNA test tell me if I am a Brahmin?

No. DNA cannot identify or prove a caste. Caste is a social identity and genes carry no jati label. A test compares your DNA with reference groups and reports how similar you are to each, which is a similarity comparison, not membership. A 2021 study even found Goan and Mangalorean Catholics genetically closest to Brahmin groups despite a different identity today.

Which Brahmin reference groups does Helixline compare against?

Our 140+ South Asian community reference groups include Kashmiri Pandit, Saraswat, Mohyal, Uttar Pradesh, Rajasthani, Maithil, Bihar, Bengali, Odia, Assamese, Nepali (Bahun), Chitpavan, Deshastha, Konkani, Goud Saraswat, Iyer, Iyengar, Namboothiri, Telugu and Kannada Brahmin groups, plus Bhumihar. Results are similarity comparisons, never a caste label.

Want to see your own proportions? Order the Decode kit in India, or upload your existing raw data from abroad.