மரபணுத் தளம்The Genome Floor
The four cited papers together form the genetic-anchor of the node argument. All are published in journals of record (BMC Genetics, Nature, Cell, Science). Each is Tier-A.
A commissioned meta-analysis with a Sri Lankan-Tamil sample expansion (currently under-represented in the reference panels) tightening the Ranaweera 2014 result; open-source ADMIXTURE runs re-published so any peer can verify.
Ranaweera et al. 2014 — the island result
A study of 476 individuals across all Sri Lankan communities using 655K autosomal SNPs, published in BMC Genetics, concluded that Sri Lankan Sinhalese, Sri Lankan Tamils, and Sri Lankan Moors cluster together and are closer to one another than any is to mainland South-Indian or North-Indian reference populations [01]. The founding-population component north-Indian-Vijaya predicts is not visible at the resolution the SNP panel can detect.
Reich et al. 2009 — the Ancestral North / South Indian model
The Nature paper 'Reconstructing Indian Population History' [02] established the ANI/ASI model. Every modern South-Asian population is a mixture of an Ancestral North Indian component (Steppe + Iranian farmer + IVC) and an Ancestral South Indian component (Onge-related + IVC-related). Tamil populations sit at the ASI-heavy end of the cline; the model is now the standard reference.
Narasimhan et al. 2019 — the timing of steppe entry
The Science paper 'The formation of human populations in South and Central Asia' [03] dates the entry of steppe (Yamnaya-related) ancestry into South Asia to after c. 2000 BCE — post-dating the mature phase of the Indus Valley Civilisation. Any 'Indo-Aryan founding' argument must sit inside this timing.
Shinde et al. 2019 — the IVC individual
The Cell paper 'An Ancient Harappan Genome Lacks Ancestry from Steppe Pastoralists or Iranian Farmers' [04] analysed a single high-coverage genome from Rakhigarhi (Indus Valley). It carries no steppe ancestry and no Iranian-farmer ancestry — a result that, combined with the ASI-cline result, is the most parsimonious floor for reading the Indus Valley as ancestrally Proto-Dravidian.
Reading the four together
Any modern claim that a specific ethnic community on the island is descended from a distinct North-Indian founding wave must show that wave in the DNA. It is not there at the resolution the panels can detect. This does not prove the Vijaya narrative false as literature — it proves it inert as biology.
BMC Genetics, Nature, Cell, Science — where any re-analysis lands, not on social media.
Institutional partners for any expanded-sample re-run. Ranaweera 2014 was Sri-Lanka-led.
Genetic-non-distinctiveness is admissible as background evidence in Art 27 ICCPR / CERD proceedings when a state claim asserts ethnic-founding priority.
- [01]Ranaweera, L. et al., 'Mitochondrial DNA history of Sri Lankan ethnic people: their relations within the island and with the Indian subcontinental populations', BMC Genetics 15:52 (2014); and companion autosomal work by the same group.
- [02]Reich, D., Thangaraj, K., Patterson, N., Price, A.L. & Singh, L., 'Reconstructing Indian Population History', Nature 461 (2009), 489–494.
- [03]Narasimhan, V.M. et al., 'The formation of human populations in South and Central Asia', Science 365:6457 (2019), eaat7487.
- [04]Shinde, V. et al., 'An Ancient Harappan Genome Lacks Ancestry from Steppe Pastoralists or Iranian Farmers', Cell 179:3 (2019), 729–735.e10.
- · Does not claim 'Tamils are more indigenous than Sinhalese'. The four papers together show high shared ancestry; the argument is that neither community is a North-Indian founding lineage in the way the Mahāvaṃsa narrative implies.
- · Does not claim IVC = Tamil. It claims IVC ancestry is most parsimoniously ancestral to modern Dravidian speakers; the linguistic bridge (Parpola, Southworth) is separate scholarship and treated as such.
- · Does not use genetics as an identity court. Identity is civic, linguistic, and self-ascribed; genetics is used here only to falsify a specific ethnic-founding claim.
- · Does not deploy genomic data on individuals — only published population-level results.