Sinhala DNA results - 2019 update

23andMe has by far the most specific and up to date DNA breakdowns for all South Asian populations.

With a reference panel of over 83,000 South Asians and growing, there is no recent DNA study that can compare.

The results of 23andMe corroborate the following 2015 study:

The following Sinhala results all show a predominant South Indian ancestry, but with significant North Indian admixture (mainly Bengali). Four Bengali results are also included for comparison:

The vast majority of these 23andMe samples are low country Sinhalese, particularly with Portuguese surnames, suggesting that they are coastal castes.

Some of the samples are low country Govigama who are also predominantly South Indian in DNA.

They differ markedly from typical North Indian and Bengali results.

2023 UPDATE

Consistent with the results of 23andMe, a 2023 study by Singh et al has also found a predominant South Indian ancestry amongst the Sinhalese with a minority of north Indian ancestry. This study used higher resolution markers than previous studies for greater accuracy.

23andMe examples

(1) Perera 1:

(2) Wijewardene:

(3) Karunathilaka:

(4) Dias:

Bengali ancestry of Sri Lankan Tamils

SL Tamil y-chromosomes from the 1000genomes project have been uploaded to YTree, an internationally recognised Y-DNA phylogenetic tree.

Some new, unique subclades have been discovered as a result, which prove paternal ancestry among the SL Tamils from both North and South India.

The most interesting haplotype discovered is J-Z7255, an unique subclade that has only been found in Bengal and Sri Lanka:


The time to most recent common ancestor (TMRCA) of all three individuals has been calculated to around 1000 B.C. This would place it in Bengal's protohistoric period (ancient Vanga).

How did a Bengali Y-chromosome end up in Jaffna?

The most likely explanation is this SL Tamil had a Sinhala paternal ancestor who assimilated into the Tamil identity during the South Indian conquests of North-East Sri Lanka.

Interestingly, the myth of Bengali prince Vijaya migrating to Sri Lanka has been preserved in Sri Lankan Tamil folktales and chronicles, but in a garbled form.

Prof. Indrapala, the foremost expert on the history of SL Tamils states the following:

"The manner in which the Sinhalese legend came to assume this position in the traditional history of the Tamils may not be difficult to explain.

The Sinhalese of the Jaffna district as we have already seen, were at no time completely dislodged by the Tamils. Many of them probably became assimilated to the Tamil population in due course. The story of Vijaya would have been current among these people at the time of Tamil settlements. When the Sinhalese became assimilated into the Tamil population, a garbled version of the Vijaya legend would have still lingered in their memory.

At at time when their origins were forgotten, these people may have used the legend to explain the origin of the Tamil kingdom instead of that of the Sinhalese kingdom."

Other unique haplotypes shared by SL Tamils also suggest paternal ancestry from ancient Gujarat and Kerala:

Both these unique haplotypes trace their time to most recent common ancestor (TMRCA) to 700 B.C.  The migration of men carrying these unique haplogroups into Sri Lanka could have occurred at any time period after that.

Vedda DNA ancestry

The Veddas are the indigenous people of Sri Lanka, and are directly descended from Balangoda man and other mesolithic hunter gatherers.

However, the Veddas are not purely indigenous and have mixed extensively with West Eurasian enriched populations over the last 3000 years (both Dravidian and Indo-Aryan speakers). 

This is proven by the below chart.

The red bar correlates with indigenous south asian DNA. The dark purple and green bars correlate with West Eurasian DNA. The yellow bar correlates with East Asian DNA.

Source: Downie 2016

1000genomes - Sri Lankan Tamils in the UK (STU) data

The 1000 Genomes Project (abbreviated as 1KGP), launched in January 2008, was an international research effort to establish by far the most detailed catalogue of human genetic variation.

Scientists sequenced the genomes of at least 1000 anonymous participants from a number of different ethnic groups including over 100 Sri Lankan Tamils in the UK (STU).

Initially, over 100 Sinhalese in the UK were also have meant to have been sampled, but for reasons unknown this was cancelled.

The data from the 1000genomes project shows significant genetic variation within the Tamil samples (plotted in black on the below chart), with some individuals clearly being of Dalit background (labelled below as S.C - scheduled caste and represented as a black triangle).

Plot courtesy of geneticist Razib Khan. Annotations added. 

(As you go up the cline, the amount of indigenous South Asian hunter gatherer DNA increases. The Bangladeshis are shifted to the right of the cline due to East Asian admixture.)

This furthers highlights the need for future genetic studies on Sri Lanka to take into account caste and region, as there is much genetic variation within the same ethnic group. Taking a single average of all these divergent castes hides an underlying complexity.

Average results from 1000 genomes project. Courtesy of geneticist Razib Khan.

Sinhalese - The Bengali and Tamil connections

The Mahavamsa, the great chronicle of Sri Lanka was written in the 5^th century AD. It recounts the migration of Prince Vijaya and his male followers from Bengal to Sri Lanka circa 500 BC.

The Mahavamsa further states that Vijaya and his followers married noble maidens from southern Madura, in the Tamil Pandyan kingdom, thus making the Sinhalese a mixture of North and South Indians from the very beginning.

This mix has been confirmed by multiple DNA studies.

Three studies indicate a predominant Bengali descent with significant Tamil admixture:

(1) The Mystery of Sinhalese Origins: An Alu perspective - S. Mastana (2007)

An Alu polymorphism analysis of Sinhalese from Colombo in 2007 using Tamil, Bengali, Gujarati (Patel) and Punjabi as parental populations found different proportions of genetic contributions:

Statistical Method	Bengali	Tamil	North Western
Point Estimate	57.49%	42.5%	-
Maximum Likelihood Method	88.07%	-	-
Using Tamil, Bengali and North West as parenteral population	50-66%	11-30%	20-23%

(2) Population genetic study of three VNTR loci (D2S44, D7S22m and D12S11) in five ethnically defined populations of the Indian subcontinent. - Papiha SS (1996) 

An VNTR study from 1996 also found over 70-82% of Sinhalese genes to originate from Bengali admixture:

Parenteral population	Bengali	Tamil	Gujarati	Punjabi
Using Tamil and Bengali as parenteral population	70.03%	29.97%	-
Using Tamil, Bengali and Gujarati as parenteral population	71.82%	16.38%	11.82%
Using Bengali, Gujarati and Punjabi as parenteral population	82.09%	-	15.39%	2.52%

(3) The Legend of Prince Vijaya - A Study of Sinhalese origins. R.L. Kirk (1976) 

"The genetic distance study agrees with this in giving low values for the distance between the Sinhalese and Tamils and Keralites.

The lowest distance value for the Sinhalese however, for fθ is with the Bengalis, but the differences in genetic distance between the Sinhalese and Tamils on the one hand and the Sinhalese and Bengalis on the other are not great. 

Indeed Bengalis are closer to the south Indian populations included in these comparisons than they are to the other Indian populations in the north and north-west."

Two other studies suggest a predominant South Indian descent with significant Bengali admixture:

(1) Genetic affinities of Sri Lankan populations. Kshatriya GK, Human Biology (1995).

"The study of genetic admixture revealed that the Sinhalese of Sri Lanka have a higher contribution from the Tamils of southern India (69.86%) compared with the Bengalis of northeast India (25.41%), whereas the Tamils of Sri Lanka have received a higher contribution from the Sinhalese of Sri Lanka (55.20%) compared with the Tamils of India (16.63%)."

(2) Genetic relationships between Indians and their neighbouring populations. Roychoudhury AK (1985)

The differences between the studies can perhaps be explained by caste and regional variations amongst the Sinhalese.

Regardless, the above genetic evidence all confirm the migration of men from the Bengal region and the mixing of North and South Indians in the early formation of the Sinhalese. 

The below chart suggests that on average the Sinhalese lie in between North and South Indian populations genetically:



Interestingly, both the Sinhalese and SL Tamils are partially enriched with the Aryan R1a1 y chromosome. This enrichment is absent in Tamil Nadu castes with the exception of the Brahmins and Jains. 

Possible Sinhala paternal ancestry of a segment of the SL Tamils may partly account for this R1a1 enrichment. 

Source: Mustak 2019

Source: Kivisild 2003

Source: Kivisild 2003

Source: Kivisild 2003

Source: Poznik 2016

Common ancestry of Sinhalese and Sri Lankan Tamils - DNA proof

"West Eurasian haplogroups among the Sinhalese, Sri Lankan Tamils and Vedda population groups with a longer history in the island suggest early migration of women carrying these haplogroups into the country. Our data led us to conclude that contemporary Sri Lankans share very close maternal ancestors and that ethnicity is created by linguistic, religious and cultural differences rather than by genetic differences." 

- Prof. Kamani Tennakoon, University of Colombo.

(This article is authored by the admin of this blog and not by Prof. Kamani Tennakoon whose work amongst other researchers is quoted from.)

Most maternal DNA studies comparing both the Sinhalese and Sri Lankan Tamils show no large genetic difference, suggesting that both populations have a common maternal ancestry.

Indeed, it is likely that Dravidian speakers who spread megalithic culture to Sri Lanka circa 1000 BC were Prakritised centuries later, along with the Vedda population.

The following studies all support this conclusion, with some even suggesting closer genetic relationships between up country Sinhalese and Sri Lankan Tamils, than with low country Sinhalese.






(1) Mitochondrial DNA history of Sri Lankan ethnic people: their relations within the island and within the Indian subcontinental populations. Lanka Ranaweera at al. (2014)
https://www.nature.com/articles/jhg2013112

"There is no clear genetic separation based on the PCA map between Sinhalese and Tamils, and between Up- and Low-country Sinhalese of Sri Lanka."

Fig 3. Principal component analysis (PCA) map of the 21 Sri Lankan subpopulations based on net genetic distances derived from haplogroup distribution frequencies.

Low-country Sinhalese (SL) : SL-Thu (Thulawelliya), SL-Lan (Lankagama), SL-Ban (Bandaraduwa).

Up-country Sinhalese (SU) : SU-Mee (Meemoure), SU-Bam (Bambarabadda), SU-Mul (Mulgama), SU-Thu (Thuppitiya), SU-Kuk (Kukulapola).

Sri Lankan Tamils (TS) : TS-Jaf (Jaffna), TS-Bat (Batticaloa), TS-Tri (Trincomalee), TS-Vau (Vavuniya).

Indian Tamils (TI)

Veddas (VA)

"The majority of Sinhalese and Tamil subgroups form close genetic proximities among themselves on both PC axes. Major exception to this clustering is found in SU-Thu. It was evident that Up-country Sinhalese are genetically closer to Sri Lankan Tamils."

(2) A study of genetic polymorphisms in mitochondrial DNA hypervariable regions I and II of the five major ethnic groups and Vedda population in Sri Lanka. Ruwandi Ranasinghe, Kamani H. Tennekoon et al. Legal Medicine (2015).

https://www.sciencedirect.com/science/article/pii/S1344622315000693?via%3Dihub

"Sinhalese, Sri Lankan Tamils and Vedda populations had a considerable presence of West Eurasian haplotypes. The Vedda population clustered separately from other groups and Sri Lankan Tamils showed a closer genetic affiliation to Sinhalese than to Indian Tamils."



Figure 1. The sharing of maternal ancestry of Sri Lankan populations in comparison with different states of Southern India.

https://www.nature.com/articles/jhg2013122.pdf?origin=ppub








(3) Genetics of Growth, Development and Human Migration - Kamani H Tennekoon (2017)

http://repo.jfn.ac.lk/med/handle/701/1590

"West Eurasian haplogroups among the Sinhalese, Sri Lankan Tamils and Vedda, population groups with a longer history in the island suggest early migration of women carrying these haplogroups into the country. Our data led us to conclude that contemporary Sri Lankans share very close maternal ancestors and that ethnicity is created by linguistic, religious and cultural differences rather than by genetic differences."

(4) Genetics profile of 11 autosomal STR loci among the four major ethnic groups in Sri Lanka - Ruwan J. Illeperuma et al.

https://www.ncbi.nlm.nih.gov/pubmed/19414153

"Allele frequencies and statistical parameters of forensic interest are presented for 11 autosomal microsatellites of four ethnic groups in Sri Lanka. A total of 513 unrelated individuals from Sinhalese, Sri Lankan Tamil, Indian Tamil and Sri Lankan Moor population groups were included."

"Our study further suggests that there is no significant genetic variation among the major ethnic groups in Sri Lanka."

(5) Genetic variants in the cytochrome P450 2D6 gene in the Sri Lankan population. Tharanga TD et al. (2013)

http://www.academia.edu/14970400/Genetic_variants_in_the_cytochrome_P450_2D6_gene_in_the_Sri_Lankan_population

"The *3 allele is absent in South Indian Tamils, the presence of the *3 allele among a large proportion of Sri Lankan Tamils to the point where it is almost reaching the high frequency observed in Sinhalese..."

"This is supported by the results of blood group genetic marker analysis, which also have not revealed any difference between the Sinhalese and Tamils."

"In summary, the most prevalent allele in the Sri Lankan population is the loss of function *3 allele, which is not present in the South Indian population."

(6) Genetic affinities of Sri Lankan populations. Kshatriya GK, Human Biology (1995)

https://www.ncbi.nlm.nih.gov/pubmed/8543296

"The study of genetic admixture revealed that the Sinhalese of Sri Lanka have a higher contribution from the Tamils of southern India (69.86%) compared with the Bengalis of northeast India (25.41%), whereas the Tamils of Sri Lanka have received a higher contribution from the Sinhalese of Sri Lanka (55.20%) compared with the Tamils of India (16.63%)."












































(7) Study of morphology, morphometry and mitochondrial DNA polymorphism of prehistoric skeletal remains of Potana and Purana population in Sigiriya, Sri Lankan. K.M. Chandimal (2014).

http://dr.lib.sjp.ac.lk/handle/123456789/5110

"Historical records indicate that the ancestry of the Purana (old) population living in the suburbs of Sigiriya in Sri Lanka can be traced back to the times of the Sinhalese kings of the 5th century A.D. (1450 YBP)...

This study was carried out to investigate maternally inherited Mitochondrial DNA (mtDNA) of the Purana population in the suburbs of Sigiriya. Forty four Purana inhabitants belonging to Purana pedigrees were recruited in this study..."

"Phylogenetic analysis based on mtDNA HVC - I polymorphism data revealed that the Purana population in Sigiriya was genetically closer to Sri Lankan Tamils than the Vedda and other modern Sri Lankans"

http://dr.lib.sjp.ac.lk/bitstream/handle/123456789/4368/Evidence%20of%20Restricted%20Maternal%20Gene%20Flow%20of%20Purana%20%28Old%29%20Population%20in%20the%20Suburbs%20of%20Sigiriya%2c%20Sri%20Lanka.pdf?sequence=1&isAllowed=y



(8) Similar Mitochondrial DNA haplotype distributions in both Sinhalese and SL Tamils strongly suggest a common maternal ancestry.

The distribution of mtDNA haplotypes among native Sri Lankans is distinctly different from both Bengalis and Indian Tamils in Sri Lanka.

Sources: Rishishwar 2017, Ranaweera 2014, Ranasinghe 2015.

Sources: Ranaweera 2014, Ranasinghe 2015 

Source: Mustak 2019

mtDNA haplotype distribution among Indian Tamils in Sri Lanka, West Bengalis and Keralites:

Source: Kivisild 2013

Source: Ranaweera 2014

Source: Mustak 2019

Harappa World Sri Lankan DNA Admixture

The Harappa DNA Ancestry project which ran from 2011-2016 has made many important discoveries including the clear DNA stratification of caste.

The Harappa World calculator which is available publicly on the Gedmatch website closely mirrors the latest South Asian DNA studies which models all South Asians primarily as a mixture of three core populations:

(1) Aboriginal hunter gatherers (AASI - Ancient Ancestral South Indians), the indigenous people of the subcontinent, and the component that makes South Asians genetically distinct from the rest of the world.

(2) Neolithic Iranian farmers (who introduced agriculture before 5000 BC and mixed with the AASI to form the Indus Valley Civilisation. These farmers likely introduced the Proto-Elamo-Dravidian languages).

(3) Bronze age Steppes pastoralists (The original 'Aryans' who entered the subcontinent circa 1500 BC).

The Neolithic Iranian farmers and Steppes pastoralists are both West Eurasian populations.

The Genomic Formation of South and Central Asia (2018)



This is a more accurate, scientific representation of the individual breakdown of South Asian populations, which in the past were erroneously described genetically as linguistic groups (e.g. racially Indo-Aryan or Dravidian, linguistic terms that should not be directly equated with genes on a one to one basis).

Harappa World DNA categories 

The seven major categories in the Harappa World calculator are the following:

South Indian
Baloch
Caucasian
NE Euro
SE Asian
Siberian
NE Asian.

These are not real populations, but the labels tell you which region these components are modal. For example, the “South Indian” component peaks in south Indian Dalits and tribals and closely correlates with the indigenous AASI component (hence why it is called 'South Indian').

The “Baloch” peaks among the Baloch people of southeastern Iran and southwest Pakistan and correlates somewhat with the Neolithic Iranian farmer component.

The “NE Euro” peaks among the eastern Baltic peoples and correlates with the Steppes people (the original Aryans).

The last three are East Asian components, running the latitude from south to north to center. They only concern populations on the eastern fringe of the subcontinent such as Bengalis.

Ancient Steppes DNA (The original Aryans) 

The following chart shows the breakdown of the DNA recovered from the skeletons of ancient Steppes people (which can be found on GEDmatch).
As it can be seen the ancient Steppes pastoralists who conquered much of Eurasia with the invention of the horse and chariot, and spread their Indo-European languages are around 60% North East European in DNA.

This is in sharp contrast to modern Iranian and Indian populations who only have a minority of Steppes DNA, despite them mainly speaking Aryan languages:

This is due to the process of elite dominance and language replacement. The Steppes people who were much smaller in number conquered these agricultural societies and replaced the local languages, whilst only contributing a relatively small amount to the gene pool.

Indeed, it is a fallacy to describe any population of modern day Iran or South Asia as being Aryan in race. They are all genetically far closer to the prehistoric farming communities that predated the Steppes migrations.

South Asian DNA

The limited data from the Harappa World calculator suggests that the Sinhalese are genetically closest to Tamils and other South Indian middle castes. These geographically close groups lie in the middle of the ‘Indian cline‘ in between South Indian Dalits and North Indian upper castes. 

The proportion of indigenous AASI genes generally rises as you go down the caste system and down the Indian subcontinent.

The South Indian Brahmins migrated from North India hence their lower proportion of ‘South Indian’ genes and higher Steppes component (North East European).

The Bengalis are clearly enriched with 'East Asian' and 'North East European' ancestry which is largely absent in Sri Lankan populations (with the exception of those recently mixed with Malays and Burghers).

This is corroborated by another recent study from 2017 which showed significant East Asian ancestry amongst Bengalis (in both the maternal and paternal lines) that is absent in most Sri Lankans:



A further study from 2015 analysed the overall autosomal DNA of 200 Sinhalese, 200 Indian Tamils, 103 Sri Lankan Tamils, 15 upper caste West Bengalis and 24 upper caste South Indians (likely Brahmins).

It found that on average the Sinhalese were genetically in between North and South Indian populations:

Harappa World DNA analysis of Sinhalese and Tamils:

https://sbarrkum.blogspot.com/2013/04/sinhalese-and-tamil-dna-admixture.html

References

1.  The Genomic Formation of South and Central Asia - Vagheesh M Narasimhan et al (2018). Available URLs:

https://www.biorxiv.org/content/10.1101/292581v1

https://www.biorxiv.org/content/biorxiv/early/2018/03/31/292581.full.pdf

2. Aryan migration: Everything you need to know about the new study on Indian genetics.
https://scroll.in/article/874102/aryan-migration-everything-you-need-to-know-about-the-new-study-on-indian-genetics

3. The mystery of the Sintashta people (Proto Indo-Iranians)
https://eurogenes.blogspot.com/2018/04/the-mystery-of-sintashta-people.html?fbclid=IwAR060dhe1ggEetUBycZGy7aqf7sY9d7TUHexXvKuo9DAaakWFD96XUDk4_g

4. Gedmatch database
https://www.gedmatch.com

5. Harappa DNA Ancestry project
https://www.harappadna.org/2012/05/harappaworld-admixture/

6. A genetic chronology for the Indian Subcontinent points to heavily sex-biased dispersals. Marina Silva et al. (2017)
https://bmcevolbiol.biomedcentral.com/articles/10.1186/s12862-017-0936-9