Rakhigarhi, the Largest Known Harappan Site, Gave the Civilisation Its First Ancient Genome in 2019, From the One Skeleton Among 61 Sampled Whose DNA Survived

On 5 September 2019 the journal Cell published an ancient genome from the Indus Valley Civilisation, taken from a woman buried in the cemetery at Rakhigarhi in Haryana. The team had sampled 61 skeletons, and only hers held enough authentic DNA to analyse. She turned out to be a mixture of people related to ancient Iranians, who made up the largest share, and South Asian hunter-gatherers, with no detectable ancestry from Steppe pastoralists, the herders of the Eurasian grasslands whose descendants reached South Asia later.

Indian institutions did the fieldwork and the sampling. According to the paper’s methods section, a team led by Vasant Shinde of Deccan College in Pune excavated the skeletons, and the ancient DNA group led by Niraj Rai at the Birbal Sahni Institute of Palaeosciences in Lucknow sampled them. David Reich’s laboratory at Harvard Medical School analysed the data. Kumarasamy Thangaraj of the CSIR Centre for Cellular and Molecular Biology in Hyderabad is one of the 28 authors.

A city of seven mounds, 150 km from Delhi

Rakhigarhi sits about 150 km from Delhi, in what the 2018 excavation paper calls an agricultural field. The Cell paper describes it as the largest known site of the civilisation, spread over 550 hectares, and the Ministry of Culture gave the same figure in a press release of 22 June 2026. Mounds 1 to 6 hold the city and Mound 7 holds the cemetery, according to the 2018 paper. For a sense of how Haryana fits into the larger map, see our profile of the state.

The Archaeological Survey of India (ASI) dug there for three years from 1997 to 2000 and uncovered a road, a drainage system and a large rainwater storage facility, the 2018 paper reports. Shinde’s Deccan College team then worked on the cemetery from 2013 to 2016. Before that, only 11 burials in the cemetery had been excavated, and the Deccan College team recorded 53.

The Cell paper says seven charcoal samples from depths of 9 to 23 metres in the habitation area give point dates between 2800 and 2300 BCE, mostly within the mature phase of the civilisation. The Ministry of Culture says excavations show continuous habitation from the Early Harappan to the Mature Harappan periods. The drains and street plan echo the ones in our piece on how Indus cities handled water.

Sixty-one samples, one genome

Ancient DNA does not keep well in heat. Narasimhan, then a Harvard Medical School postdoctoral fellow and a co-first author, told Live Science that the hot, humid climate of the region degrades it quickly. Nick Patterson, a co-author, told Scroll that “two words I would not use to describe India are cool and dry.” Live Science reported the result as the first time an individual from the civilisation had yielded any DNA information, and Nature India called it the first genome sequence of a woman from Rakhigarhi.

The cemetery lies about 1 km west of the ancient town. The team made 251 DNA libraries from 61 samples, and 108 of those came from a single skeleton, catalogued as I6113. The Harvard Medical School write-up says that took more than 100 sequencing attempts. The longer DNA fragments showed signs of modern contamination, so the team kept only fragments carrying the chemical damage typical of ancient DNA. That left 31,760 positions in her genome to analyse.

The skeleton is female, which the ratio of Y to X chromosome sequences confirms. Her mitochondrial DNA, inherited only through the mother, belongs to haplogroup U2b2, which the paper says is absent from data on about 400 ancient Central Asians and is today almost exclusive to South Asia.

Radiocarbon dating failed. The bones kept too little collagen, and five attempts on her skeleton gave unusable carbon-to-nitrogen ratios. The authors date the grave from context: Rakhigarhi shows only Harappan occupation, the cemetery follows Harappan burial customs, and the pottery buried with her is in the style of the mature phase.

What the genome shows

The paper models her ancestry as Iranian-related plus a South Asian hunter-gatherer component, for which the Andamanese serve as the nearest sampled stand-in. In one version of the model the Iranian-related share is 73 per cent, give or take 6, and in another it is 63 per cent, give or take 6. “Iranian-related” is a label for a genetic lineage and says little about geography. Reich told Nature India, “We say ‘Iranian-related’ because we don’t know where they lived,” and added that the team’s data point to her ancestors living in South Asia for many thousands of years before the civilisation began.

Her profile matches 11 people buried at Gonur in Turkmenistan and Shahr-i-Sokhta in eastern Iran, both towns in contact with the Harappans. The authors give three reasons to think the genes travelled out of South Asia. Only 11 of 44 good-quality individuals at those two sites carry the profile. The three Shahr-i-Sokhta individuals with material links to Baluchistan all carry it. And the hunter-gatherer share in them is absent from the other Shahr-i-Sokhta samples and matches ancestry found in South Asians. The paper supports the conjecture that those 11 were migrants from the Harappan world, and renames the group of 12 the IVC Cline, a gradient of ancestry across the civilisation’s individuals.

Little or no Steppe pastoralist ancestry shows up in the 12, which the authors say shows it was not common in north-west South Asia during the civilisation as it is today. They place the arrival of Steppe ancestry in South Asia at about 2000 to 1500 BCE. They also say the profile of the Rakhigarhi woman is no longer found among South Asians or Iranians today, and that gene flow reached the region from the north-west and the south-east after her time.

On farming, the paper finds that the Iranian-related lineage in these people split off earlier than a hunter-gatherer from Belt Cave in the Alborz mountains, about 10000 BCE, and the goat herders of Ganj Dareh in the Zagros mountains, about 8000 BCE. It finds negligible ancestry related to Anatolian farmers, so the authors conclude that the first farmers of the Fertile Crescent did not bring farming to South Asia. Narasimhan told Live Science the lineage “diverges at a time prior to the advent of farming almost anywhere in the world.”

Reich added a caution. He told Nature India the findings “do not prove a separate invention of farming in South Asia.” Partha Pratim Majumder of the National Institute of Biomedical Genomics added in the same report that fossil pollen points to plants being domesticated several times in India.

The companion paper in Science, with Narasimhan as lead author and 523 ancient people in its dataset, reports that the primary source of ancestry in modern South Asians is a gradient between people related to early hunter-gatherers of Iran and Southeast Asia. Its abstract says that after the civilisation’s decline its people mixed with others in the south-east, forming one of the two main ancestral populations of South Asia, whose direct descendants live in southern India. They also mixed with the descendants of Steppe pastoralists, forming the other.

What one woman cannot show

The paper limits its own reach. Its closing section says that “a single sample” or even the gradient of 12 samples “cannot fully characterize a cosmopolitan ancient civilization”, and it asks for DNA from more people across the civilisation’s range.

Jonathan Mark Kenoyer of the University of Wisconsin, Madison, who was not part of the research, made a related point to Live Science. He said cemetery burials are not representative of the ancient populations of the Indus cities, and that burial was less common than cremation. Narasimhan, in the same report, compared Indus cities to modern Tokyo or New York, where people gather from around the world.

The paper adds that the place where South Asian farming began is, in the authors’ words, “a yet-unsampled location”.

The Aryan argument, with names attached

The disagreement began in the days after publication. According to Scroll, Shinde put out a press release on 6 September 2019 that said the data “completely sets aside the Aryan Migration/Invasion Theory” and proves that the “Harappans were the Vedic people.” At a Delhi press conference, as ThePrint reported, he said the development once credited to outsiders was the work of indigenous people.

Rai spoke more narrowly. ThePrint quotes him saying the team had not found enough evidence to call the movement of Central Asians an invasion, and that the difference between migration and movement matters. Scroll reports that he did not agree with the claim that the Harappans were the Vedic people.

Narasimhan told Scroll that the proposition “makes jumps that I am not comfortable with.” Patterson told Scroll he disagreed that the Indus people spoke an Indo-European language. The Cell paper’s own discussion takes a position. It calls a spread of Indo-European languages from Eastern Europe through Central Asia in the first half of the 2nd millennium BCE a natural route, and says the Steppe ancestry in South Asia matches that of Bronze Age Eastern Europe. It also says its findings argue against a spread of those languages across the Iranian plateau with farming, the so-called Anatolian hypothesis.

Swarajya’s Aravindan Neelakandan wrote that the study brought to light an indigenous Harappan ancestry, and quoted the geneticist Gyaneshwer Chaubey as saying it showed farming in India did not come with Iranian input. Swarajya adds that Chaubey was cautious about drawing language conclusions from genetics. The paper’s data point is plain: the Rakhigarhi woman carried no Steppe ancestry, and the paper dates the arrival of that ancestry to after her time. What the data say about language and the Vedic texts is what the scholars above dispute.

Where Dholavira and Lothal fit

The Cell paper calls the Harappan world cosmopolitan, with multiple cultural groups living together in large centres. It names Harappa, Mohenjo-daro, Rakhigarhi, Dholavira in Kutch and Ganweriwala. Its only genome from inside that world comes from Rakhigarhi, so the paper reports no data from Dholavira or Lothal.

Our piece on Dholavira covers its reservoirs and the board of ten signs nobody can read. The one on Lothal covers the sealings and bead workshops that scholars use to argue for trade. The Cell paper adds people to that picture: 11 individuals in Iran and Turkmenistan whom the authors read as migrants from the civilisation.

The skeletons now in Kolkata

The ASI’s Excavation Branch-II at Greater Noida worked at Rakhigarhi in the 2025 to 2026 season and found eight burials on Mound 7, the Ministry of Culture says. Three complete skeletons and fragments from other burials have been handed to the Anthropological Survey of India’s laboratory in Kolkata, under a memorandum of understanding between the two bodies. The remaining material was expected within a few days of the release of 22 June 2026.

The Anthropological Survey says the work will be done with the Birbal Sahni Institute, University College London and a Banaras Hindu University team on ancient DNA, and will include isotope studies and examinations of the bones. Thangaraj said ancient DNA on the remains could show how human genomes evolved, adapted and underwent natural selection since around 3000 BCE, and Chaubey called the collaboration a major step forward. Modern genomes are a separate project, covered in our piece on GenomeIndia. As of 15 October 2026 we found no published genome from any second Rakhigarhi burial, and the release gives no date for results.