East African Indian DNA: Ancestry for Families from Kenya, Uganda and Tanzania

Ask a family in Leicester, Wembley, Toronto or Vancouver where they are from, and the answer often comes in two parts. "We are Gujarati, but we came from Kampala." "My grandfather was born in Nairobi, his father came from a village in Kutch." "Dad grew up in Dar es Salaam, Mum in Mombasa, and their parents were from Punjab and Goa."

These are the families that historians and sociologists often call twice migrants: people of Indian origin whose families first settled in East Africa, in Kenya, Uganda, Tanzania, Zanzibar or Malawi, and who then moved again, mostly to Britain and Canada and later to the United States. Many of them have taken a consumer DNA test and found a result that feels oddly flat: "Broadly South Asian", or a single large block labelled "India", with no sign of the century their family spent in Africa.

This guide explains why. It walks through the history that took Indian families to East Africa, what peer-reviewed genetics says about Gujarati and Punjabi ancestry, what a DNA test can and cannot tell a twice-migrant family, and how to get a more detailed picture from the raw data file you already own.

Who are the East African Indians?

"East African Asians" is the label used in Britain for people of South Asian origin who came from Kenya, Uganda, Tanzania (including Zanzibar) and neighbouring countries such as Malawi. In practice it covers many different communities, with different languages, faiths and histories, who happened to share a region and, later, a second migration.

A useful snapshot comes from Uganda in 1972. Writing for the Canadian Museum of Immigration at Pier 21, the historian Jan Raska describes the Asian population of Uganda at that time as more than 80,000 people, mostly of Indian or Pakistani origin. He gives a rough breakdown: about half were Gujarati Hindus, about 30 percent were Ismaili Muslims, and the remaining fifth included Sikhs, Goans, Punjabi Hindus, Ithna Asharis, Bohras and a few Parsis.

That mix is a good guide to the wider region. Gujarati speakers, from mainland Gujarat, Kathiawar (Saurashtra) and Kutch, formed the largest share, and they themselves were divided into many communities: Hindu trading and farming communities such as Lohanas and Patels, Jain families such as the Oshwals, and Muslim communities such as the Ismaili Khojas, Bohras and Memons. Punjabi Sikhs and Hindus formed a second major strand, and Goan Catholics a third, often working as clerks, teachers and civil servants.

None of these communities is uniform. Each contains families from different villages, with different marriage histories and different routes to Africa, which is a reason to be careful with any claim that one test can define a whole community.

Why "twice migrants"?

The phrase captures two departures: from India to a British-ruled East Africa, often generations ago, and then from East Africa, often suddenly and under pressure, to Britain, Canada, the US and elsewhere. The genetic story mostly reflects the first departure. The cultural story reflects both.

A history in four waves: from dhows to Heathrow

1. Centuries of Indian Ocean trade

Contact between western India and the Swahili coast is very old. Monsoon winds made the crossing between Gujarat, Kutch, Sindh, Arabia and East Africa practical for sailing ships, and merchants used them for centuries. A 2023 study in Nature by Brielle and colleagues sequenced ancient DNA from 80 people buried in medieval and early modern Swahili coastal towns (AD 1250-1800). It found that many of them had a large share of Asian ancestry alongside African ancestry, that this Asian ancestry included components associated with both Persia and India, and that most of it came from Persian men. The authors date the start of this mixing to about AD 1000. That is a finding about the Swahili coast population, not about the Indian families who arrived much later, but it shows how long these shores have been connected.

By the 19th century, Kutchi merchants were central to the trade of Zanzibar and Muscat. The historian Chhaya Goswami's book The Call of the Sea: Kachchhi Traders in Muscat and Zanzibar, c. 1800-1880 (Orient BlackSwan, 2011) uses archival sources to show how these traders linked Zanzibar and Muscat to Bombay and Mandvi, traded in goods such as ivory and cloves, and acted as bankers and financiers for the wider Indian merchant network. Many East African Gujarati families trace their story to this world of Kutchi and Kathiawari trading houses rather than to the railway.

2. The Uganda Railway, 1896-1901

The best known chapter is the Uganda Railway, which the British built from Mombasa on the coast, starting in 1896, to Kisumu on Lake Victoria, reached in 1901. According to the National Railway Museum in York, about 32,000 workers came from British India over the course of construction, largely Sikhs from the Punjab. The museum notes that 2,493 are reported to have died in East Africa, and that around 6,724 of them stayed on after the line was finished.

The railway matters to the story, but it is easy to overstate. Most of the workers went home. And the railway labour force was largely Punjabi, whereas by 1972 about half of Uganda's Asians were Gujarati Hindus. That suggests that much of the later community grew from traders, shopkeepers, clerks and professionals who came separately in the decades that followed, as well as from the railway workers who stayed.

3. Independence and Africanisation in the 1960s

Uganda became independent in 1962, and its neighbours Kenya and Tanganyika at around the same time. Many Asian residents held British passports rather than local citizenship, and in the years that followed, new governments introduced policies to move jobs and businesses into African hands. In Kenya, laws on work permits and trade licences in the late 1960s put intense pressure on non-citizen Asian traders and employees.

Many Kenyan Asians used their British passports to leave for the UK. In 1968 the British government rushed through the Commonwealth Immigrants Act, which restricted the right of entry of British passport holders without close ties to the UK. The historian Randall Hansen, writing in The Historical Journal in 1999, describes this as the decision to withdraw the right of entry from Asians with British passports who were being driven out of Kenya by its Africanisation policies.

4. The 1972 expulsion from Uganda

On 4 August 1972 President Idi Amin ordered the expulsion of Uganda's Asian population and gave them 90 days to leave. According to the UK National Archives, just over 28,000 men, women and children came to the UK. The National Archives' own collection of family stories mentions resettlement camps such as Greenham Common, Stradishall and Tonfanau in Wales, from which families moved on to build new lives in towns and cities across Britain.

Canada opened an office in Kampala on 6 September 1972 and organised charter flights before the 6 November deadline. Jan Raska's account for Pier 21 records that Canada granted 6,175 visas, that 31 charter flights carried 4,426 people, and that more than 7,000 Ugandan Asian refugees had been resettled in Canada by the end of 1973. Raska also notes that many Ismailis had chosen Ugandan citizenship after independence and were hard hit by its loss, and that thousands of Ugandan Asians were left stateless.

Timeline: the twice-migrant journey

When	What happened	Source
From about AD 1000	African and Asian ancestry, including India-associated components, begins to mix in Swahili coast towns	Brielle et al., Nature, 2023
c. 1800-1880	Kutchi traders link Zanzibar and Muscat with Bombay and Mandvi, acting as merchants and financiers	Goswami, The Call of the Sea, 2011
1896-1901	Uganda Railway built from Mombasa to Kisumu; about 32,000 workers from British India, around 6,724 stay	National Railway Museum
Early 1960s	Kenya, Uganda and Tanganyika gain independence; many Asians keep British passports	Pier 21 (Raska)
Late 1960s	Africanisation pressure in Kenya; UK passes the Commonwealth Immigrants Act 1968	Hansen, The Historical Journal, 1999
4 August 1972	Idi Amin orders the expulsion of Uganda's Asians, with 90 days to leave	UK National Archives; Pier 21
1972-1973	Just over 28,000 come to the UK; more than 7,000 are resettled in Canada	UK National Archives; Pier 21

If your family story fits none of these waves, that is normal. For the broader British picture, see our guide to DNA ancestry for British Indians in the UK covers the wider community.

What your DNA actually records: India, not the stopover

The most common question from twice-migrant families is whether their DNA will "show Africa". For most families the honest answer is no, and that is not a fault in the test.

DNA is inherited from parents, not from places. A family that lived in Nairobi for four generations but married within Gujarati, Punjabi or Goan communities throughout will carry essentially the same ancestry as relatives who never left India. An ancestry analysis compares your DNA with reference populations, so it will point to the regions of South Asia your ancestors came from. Residence in Kenya, Uganda or Tanzania leaves no genetic trace by itself.

When African ancestry can appear

Some families do have African ancestry, through a marriage in a particular generation, and some have Arab, Persian, European or other ancestry for the same reason. We would not assume either way for any family. If such an ancestor exists, a good analysis can usually show it, because recent ancestry from a different population tends to appear as long, continuous segments on specific chromosomes rather than scattered small fragments.

The size of the signal roughly halves with each generation: a grandparent contributes about a quarter of your DNA on average, a great-grandparent about an eighth, and an ancestor five or six generations back may leave very little that a test can detect. So a small percentage in an unexpected category should be read with caution, and a clean "100 percent South Asian" result does not rule out every distant ancestor.

The Gujarati reference sample that started in Houston

There is a neat twist in the genetics of Gujaratis. One of the most widely used Gujarati reference samples in genetics research is itself a diaspora sample. The 1000 Genomes Project, and before it the HapMap project, includes a population called GIH: Gujarati Indians in Houston, Texas. Researchers have analysed it for years as a stand-in for Gujarat.

It turns out to be more complicated than a single group. A 2016 study in Genome Biology and Evolution by Sengupta and colleagues confirmed earlier reports of population substructure within the HapMap Gujarati sample, and found similar substructure in the 1000 Genomes Punjabi and Telugu samples. The authors concluded that this structure is better explained by differences in the proportions of ancestral components and by endogamy-driven social structure than by any new ancestral source, and they cautioned that language or geography alone is an inadequate proxy for an ethnic group in the subcontinent. In plain terms: "Gujarati" is not one genetic population, and nor is "Punjabi".

Why global tests give a flat result

That is exactly why a broad label on a global test feels unsatisfying to an East African Gujarati family. Global companies built their reference panels mostly from European, African and East Asian customers, so with few well-labelled samples from specific Gujarati, Kutchi or Punjabi communities their algorithms fall back on wide labels such as "Broadly South Asian". A small "Middle Eastern" or "East African" slice of a few percent is often statistical noise, partly because ancient ANI ancestry is itself related to West Asian populations. Chromosome painting is the best check: real recent ancestry shows as long segments. Our guide to NRI and diaspora DNA testing covers this in more depth.

The deeper layers: ANI, ASI and Gujarati and Punjabi genetics

Beneath community-level differences sits a deeper structure shared by almost everyone in India. In 2009 Reich and colleagues, writing in Nature, analysed 25 diverse Indian groups and found strong evidence for two ancient populations ancestral to most Indians today: Ancestral North Indians (ANI), genetically close to Middle Easterners, Central Asians and Europeans, and Ancestral South Indians (ASI), as distinct from ANI and East Asians as they are from each other. They estimated that ANI ancestry ranges from 39 to 71 percent in most Indian groups.

A 2013 follow-up in the American Journal of Human Genetics by Moorjani and colleagues studied 73 groups and dated the mixing of ANI and ASI to between about 1,900 and 4,200 years ago. They described a shift from a time when mixture between groups was common to one in which mixture, even between closely related groups, became rare because of endogamy.

For East African Indian families this has two practical consequences:

• Gujarati and Punjabi are Indo-European languages, and Reich and colleagues found ANI ancestry was higher in Indo-European speakers, but there is real variation. Two Gujarati families from different communities, or even different villages, can show noticeably different proportions. That reflects old marriage patterns in India, not anything that happened in Africa.
• Endogamy made communities genetically distinguishable. Centuries of marrying within a community leave statistical fingerprints. That is what makes community similarity comparisons possible, and it is also why they must be read as similarity, not identity.

If you want to go deeper on these layers, read ANI and ASI: the two ancestral groups in Indian DNA, our longer guide to Gujarati DNA and ancestry, and the companion piece on Indian Ocean trade routes and DNA evidence.

Community by community: what a comparison can and cannot show

This section is a guide to which reference groups are useful for different East African Indian families. It is not a ranking, and no community is more "authentic" or "pure" than another. Every one of these communities has absorbed people, moved and changed over centuries.

Gujarati Hindu families: Lohana, Patel/Patidar and others

Lohana and Patel/Patidar families were among the most prominent Gujarati communities in East Africa. Both Lohana and Patel/Patidar are reference groups in our panel, so a Gujarati Hindu family from Kampala or Nairobi can see how closely their DNA resembles each, alongside other Gujarati groups such as Koli and Sindhi (Hindu). Our dedicated guide to Patel and Patidar DNA ancestry goes into more detail.

Jain families, including Oshwals

Jain families, especially the Halari Visa Oshwals from the villages around Jamnagar, are an important part of the East African story. According to the community's own published history, Rise and Glory: History of Halari Visa Oshwals by Rati Dodhia, a few young men sailed by dhow from Porbandar to Mombasa in 1899, others followed, and most of the community later moved on to the UK after the East African countries became independent. We do not have a separate Jain or Oshwal reference group, so a Jain family's DNA is compared with neighbouring Gujarati groups. Religion is not a genetic category, and Jain families typically resemble the regional populations they lived among in India.

Ismaili Khoja, Bohra and Memon families

Muslim Gujarati and Kutchi communities were a large part of East African Asian life. Our panel includes Dawoodi Bohra and Kutchi Memon reference groups. We do not have a separate Ismaili Khoja group, so Khoja families are compared with these and other Gujarati and Sindhi groups. Conversion changes faith, not ancestry, so Muslim Gujarati families generally show strong similarity to Gujarati and Sindhi reference groups.

Punjabi Sikh and Hindu families

For families descended from railway workers, artisans, police and traders from Punjab, the useful references include Jat Sikh, Ramgarhia Sikh, Khatri Sikh, Punjabi Khatri and Punjabi Arora groups in our panel. Many Punjabi East African families later settled in Canada and the UK; our guide to Indo-Canadian Punjabi DNA covers that side of the story.

Goan families

Goan Catholics were a distinctive part of colonial East Africa. Our panel includes Goan Catholic and Konkani Brahmin reference groups. Many Goans expect to see Portuguese ancestry; genetics usually shows a mainly Indian profile, which our article on Goan and Mangalorean Catholic DNA explains in detail.

Mixed families

Many families, especially after resettlement in Britain and Canada, married across communities. A mixed background often shows up as partial similarity to more than one group, which is exactly what you would expect.

Family background in East Africa	Useful reference groups in our panel	Worth knowing
Gujarati Hindu (Lohana, Patel and others)	Lohana, Patel/Patidar, Koli, Sindhi (Hindu)	Similarity, not membership; villages differ
Gujarati Jain (including Oshwal)	Neighbouring Gujarati groups	No separate Jain reference group
Ismaili Khoja, Bohra, Memon	Dawoodi Bohra, Kutchi Memon, Sindhi (Hindu)	No separate Khoja reference group
Punjabi Sikh or Hindu	Jat Sikh, Ramgarhia Sikh, Khatri Sikh, Punjabi Khatri, Punjabi Arora	Y-DNA and mtDNA add paternal and maternal lines
Goan Catholic	Goan Catholic, Konkani Brahmin	European ancestry is usually small or absent

What a Helixline upload adds for East African Indian families

You do not need a new kit. Your existing raw data file from 23andMe, AncestryDNA, MyHeritage or FamilyTreeDNA can be re-analysed against a reference panel focused on South Asia, built on 7,600+ global reference samples with 140+ South Asian community reference groups. The Upload Ancestry report includes:

• Community similarity comparisons against those reference groups, including the Gujarati, Punjabi and Goan groups named above, with a closest-match shortlist. These are similarity comparisons, not a verdict on identity or caste.
• A state-level ancestry breakdown, so you can see how your profile relates to Gujarat, Punjab, Goa and other regions.
• ANI and ASI composition, with the steppe and AASI ancient components, and chromosome painting, which helps you tell a real recent segment from noise.
• Y-DNA and mtDNA haplogroups, if your raw file includes those markers.

Upload Complete adds wellness traits, carrier screening, pharmacogenomics insights and downloadable PDF reports. These are informational insights, not a diagnosis.

Plan	UK	Canada	US	Other currencies
Upload Ancestry	£19	CA$34	$25	€23 · A$38 · S$34 · AED 92 · ₹1,999
Upload Complete	£39	CA$68	$50	€46 · A$76 · S$68 · AED 184 · ₹3,999

How it works

1. Choose your plan and currency on helixline.in/upload and pay once.
2. We email you a link to your dashboard.
3. Sign in with the same email and upload your unmodified .txt or .zip raw data file.
4. Your results are ready within 1-2 days.

Your data is encrypted, stored in India under the Digital Personal Data Protection Act 2023, and never sold. You can delete your uploaded raw file from your dashboard at any time. Country guides for families in the UK and families in Canada explain the options in local terms, and the upload terms cover refunds.

Testing the generation that stayed in India

Many twice-migrant families still have relatives in Gujarat, Kutch, Punjab or Goa. Testing an older relative who never left gives you a family baseline, and carries more of the DNA of the generations that crossed the ocean.

You can buy a kit from abroad for someone at an Indian address. Choose "Someone in India" at checkout on the buy page, or see our gift page for gift wrap and a printed card. The kits are:

• Origins (₹6,999): ancestry with a state-level breakdown, community similarity, ANI/ASI/steppe/AASI composition and 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.

Collection is a simple saliva sample with free doorstep pickup. Kits arrive in 3-4 days across India with free shipping. Once we receive your sample, results are available within 2-3 weeks. Kits ship within India as standard; if a relative lives in another country, delivery abroad can be arranged on request with a quote through our international page.

Reading your results as a family

Test more than one relative if you can, especially parents or grandparents from different communities or regions. Siblings share about half their DNA, so two siblings can show slightly different community similarities, and that is normal. Keep the paper trail alongside the DNA: British and East African passports, school and marriage certificates, resettlement documents and the names of villages your grandparents mentioned. DNA can help with the first migration; paper and conversation help with both.

Sources and further reading

• Reich D, Thangaraj K, Patterson N, Price AL, Singh L. Reconstructing Indian population history. Nature, 2009. pmc.ncbi.nlm.nih.gov/articles/PMC2842210
• Moorjani P et al. Genetic evidence for recent population mixture in India. American Journal of Human Genetics, 2013. pmc.ncbi.nlm.nih.gov/articles/PMC3769933
• Sengupta D, Choudhury A, Basu A, Ramsay M. Population stratification and underrepresentation of Indian subcontinent genetic diversity in the 1000 Genomes Project dataset. Genome Biology and Evolution, 2016. academic.oup.com/gbe/article/8/11/3460/2680038
• International Genome Sample Resource. Gujarati Indians in Houston, TX (GIH) population. internationalgenome.org/data-portal/population/GIH
• Brielle ES et al. Entwined African and Asian genetic roots of medieval peoples of the Swahili coast. Nature, 2023. nature.com/articles/s41586-023-05754-w
• Goswami C. The Call of the Sea: Kachchhi Traders in Muscat and Zanzibar, c. 1800-1880. Orient BlackSwan, 2011. publisher listing
• National Railway Museum. Beyond the Lunatic Line. blog.railwaymuseum.org.uk/beyond-the-lunatic-line
• Uganda Railway (construction dates and background). en.wikipedia.org/wiki/Uganda_Railway
• Dodhia R. Rise and Glory: History of Halari Visa Oshwals. 2005 (a community history). jainfoundation.in
• Hansen R. The Kenyan Asians, British politics, and the Commonwealth Immigrants Act, 1968. The Historical Journal, 1999. cambridge.org
• The National Archives (UK). Marking the 50th anniversary of the arrival of Ugandan Asians in Britain. nationalarchives.gov.uk
• Raska J. Canada's Oppressed Minority Policy and the Resettlement of Ugandan Asians, 1972-1973. Canadian Museum of Immigration at Pier 21. pier21.ca

Frequently Asked Questions

Will my DNA show East African ancestry if my family lived in Kenya or Uganda?

Usually not. DNA is inherited from parents, not from places, so a family that lived in Kenya, Uganda or Tanzania for generations but married within Gujarati, Punjabi or Goan communities will show Indian ancestry, much like relatives who never left India. If a particular ancestor married into an African, Arab or other family, a good analysis can often show it as long continuous segments on specific chromosomes. We would not assume either way for any family.

What does twice migrant mean?

Twice migrants are people of Indian origin whose families first settled outside India, in this case in East Africa (Kenya, Uganda, Tanzania, Zanzibar or Malawi), and then moved again, mostly to Britain and Canada and later the United States. Many left after the Africanisation policies of the late 1960s or the 1972 expulsion of Asians from Uganda, when just over 28,000 people came to the UK and more than 7,000 were resettled in Canada by the end of 1973.

Can DNA tell me if my family were Lohana, Patel or Oshwal?

No. DNA cannot identify or prove a caste or community. What an analysis can show is how similar your DNA is to different reference groups. Our panel includes Lohana and Patel/Patidar reference groups, along with other Gujarati groups such as Koli, Dawoodi Bohra and Kutchi Memon, but similarity is not membership. We do not have a separate Jain or Oshwal reference group, so Jain families are compared with neighbouring Gujarati groups.

Why does 23andMe call my East African Indian family Broadly South Asian?

Global companies built their reference panels mostly from European, African and East Asian customers, so they have relatively few well-labelled samples from specific Gujarati, Kutchi or Punjabi communities. Their algorithms fall back on wide regional labels. Research on the 1000 Genomes Gujarati sample (Gujarati Indians in Houston) also shows that Gujarati is not one genetic population, which is why comparisons with specific community reference groups are more informative.

How much does a Helixline upload cost in the UK and Canada?

Upload Ancestry costs £19 in the UK, CA$34 in Canada and $25 in the US, with prices in euros, Australian and Singapore dollars, dirhams and rupees as well. Upload Complete, which adds wellness traits, carrier screening, pharmacogenomics and PDF reports, costs £39, CA$68 or $50. You upload the raw file you already have from 23andMe, AncestryDNA, MyHeritage or FamilyTreeDNA, and results are ready within 1-2 days.

Can I send a DNA kit to my relatives in India from the UK or Canada?

Yes. Choose Someone in India at checkout on the Helixline buy page, or use the gift page, and the kit is delivered to their Indian address with free shipping and free doorstep pickup of the saliva sample. Origins costs ₹6,999, Decode ₹12,999 and Infinite ₹17,999. Kits arrive in 3-4 days across India, and results are available within 2-3 weeks after we receive the sample. Delivery to other countries can be arranged on request with a quote through the international page.

Curious what your own file shows beyond "Broadly South Asian"? Upload your raw data to Helixline.