PMC

The locations of the three contemporary Indian tribal populations surveyed are given on the regional map.

Each line represents the four systems of G, A, C, and T, in descending order.
SNPs are named by their position relative to the Cambridge Reference Sequence and are separated by variation in electrophoretic mobility, which is determined by the length of the SBE primer.
The segregation of the two samples at all lineage defining sites, for M31a1 and M32 respectively, demonstrates the accuracy and reliability of genotyping using this methodology.

Alleles at np15754 and 3817 on the G line are from the contaminant DNA, whilst those on the A system represent the M32 sample.
The level of the contaminant increases, as a proportion of the total mix, with each of the descending lines shown within the oval.
The relative amounts of contaminant are 4%, 8% and 20% of the effective copy number (assessed by quantitative PCR) prior to mixing of the two extracts to seed the multiplex.

A-G correspond to our hapolotypes assigned to M31a1b1, M31a1b*, M31a1*, M31a1a, M32*, M31a* and M32a1, respectively, whilst H-K are from Thangaraj et al.
The node marked “Root” is the root of haplogroup M and ticks on the network indicate mutations, the most important of which are shown by their position; the long internal branches are shown compressed.
The diagram summarizes all most parsimonious reconstructions of mutations on a tree.
Note how our data can be explained (bold) by a single homoplasy at position 8108.
However to attach H, I and J to the bold tree would require a further four recurrent mutations to be introduced.

The definitions of the Andaman-specific M31a1 and M32 lineages are revised to include additional control region SNPs resulting from the cloned sequences of the historic samples.
The Indian variant found amongst the Lodha, Chenchu and Lambadi develops the inter-regional structure of M31 and allows for the settlement of the Andaman archipelago substantially after the last successful migrations Out of Africa by modern humans.
These sequences contribute a deletion at np249 on the trunk of M31a1 and the motif of 143-195-207 on the stem of M32, contra to the tree of Thangaraj et al., which did not report the deletion and the three M32 SNPs never occurred together in their five samples.
The original data also reported a transition at np200 on all four of their M31 samples, but in the historic data it is only found in a single Onge sample.