Genetic Ancestry of the Andamanese

Conclusions
The comparative analysis of Y-chromosome sequences of
many Indian populations—with different locations, languages and
social structure—has demonstrated that Indian
populations have a complex ancestry which cannot be
explained by a single expansion from the North (or North
West) to the Indian subcontinent on a uniform pre-existing
population. It is interesting to note that Y-chromosome data
displayed less diversifcation between the North and South
than has been described in autosomal studies (Basu et al.
2003; Reich et al. 2009; Moorjani et al. 2013; Juyal et al.
2014), with no clear clines of frequency of haplogroups in
the present dataset.
The analysis of full Y-chromosomal sequences allows a
deeper analysis, with assessment of the time depth of population-
specifc clades, and of particular interest, of the time
depth and place of origin of clusters of Indian sequences
compared with the closest sequence found outside India
in a worldwide survey. The lack of a North to South cline
of frequency of any haplogroup and similar time of divergence for
both North and South India non-tribal populations suggest a similar
paternal ancestry for both of these
populations.
Our time divergence estimate matches the previous
studies which argued that most of the haplogroups present
in India (C5, F*, H, L1 and R2) arose inside India rather
than being brought from outside (Sengupta et al. 2006;
Carvalho-Silva et al. 2006). The puzzling point is that the
well-recognized later Indo-European migration, which
strongly affected the northern regions, did not produce
detectable major changes in the Y-chromosome gene-pool.
This could be explained either by a similar Y-chromosome
background in the two main migrations (for example, Neolithic
and Indo-European) or that the second one did not
have a major genetic impact, at least for the Y-chromosome
(Sengupta et al. 2006). In general, these results support the
view of Thapar (2014) that Dravidian speakers were spread
geographically throughout India and that the incursion of
Indo-Aryan speakers in the Northwest dominated pre-existing
Dravidian tribal populations, with an élite-dominance
model and small genetic impact.
Even more interesting, the closest neighbours of Indian
clades in our dataset are generally from Southern Europe
(and not other European populations), a place known to
have had more influence from the first Neolithic expansion
from the Levant through Anatolia and less from the steppe
migration which was perhaps responsible for the IndoEuropean
expansion of languages in Europe (Haak et al.
2015); the future availability of ancient Y-chromosome
sequences and reanalysis after merging available data
from Western Asia (Hallast et al. 2014; Batini et al. 2015;
Karmin et al. 2015) will help to better interpret this fnding.
The time divergence between Indian and European
Y-chromosomes, based on the closest neighbour analysis,
shows two different distinctive divergence times for J2 and
R1a, suggesting that the European ancestry in India is much
older (>10 kya) than what would be expected from a recent
migration of Indo-European populations into India (~4 to
5 kya). Also the proportions suggest the effect might be
less strong than generally assumed for the Indo-European
migration. Interestingly, the ANI ancestry was recently
suggested to be a mix of ancestries from early farmers of
western Iran and people of the Bronze Age Eurasian steppe
(Lazaridis et al. 2016). Our results agree with this suggestion.
In addition, we also show that the divergence time
of this ancestry is different, suggesting a different time to
enter India.
These results suggest that the European-related male
ancestry in Indian populations might be much older and
more complex than anticipated, and might originate from
the frst wave of agriculturists or even earlier, giving
stronger support to the very old ages for Indian Y-chromosomes;
it also downplays the importance of migration
related to the Indo-Aryan linguistic expansion. It is interesting
to note that both the closest neighbours of North and
South Indian clusters have a lowest minimum around ~4 to
5 kya coinciding with the Indo-European migration, suggesting that
the real stratifcation of these two populations
started around that time, and may be related to a retraction to the
South of Dravidian speakers (Thapar 2014).
Due to the low number of sequences from Dravidian and
Austro-Asiatic tribes, it is diffcult to be certain about the
ancestry of these tribal populations. Nonetheless, both of
these populations have shared recent ancestry (~10 kya)
with their non-tribal neighbouring populations (BIR with
North Indian and ILA with South Indian) which contrast
with that seen from autosomal data (Mondal et al. 2016).
The Andamanese and some Japanese (and also Tibetans, not present
in this data set) showed the presence of
haplogroup D Y-chromosomes, which provided one of
the main reasons for previous genetic studies to postulate
that the Andamanese represented an out-group to all other
OOA populations, descendants of an early modern human
expansion, following a coastal route migration from which
only a few populations survive to the present day; we have
been able to dismiss this hypothesis. We have shown that
Andamanese and Japanese individuals carrying haplogroup
D separated around ~ 53 kya, coinciding with most of the
other major OOA haplogroup divergences. We have also
demonstrated, using autosomal data, that the Andamanese are
indeed an out-group to East Asian populations,
which was also shown before (Chaubey and Endicott 2013;
Aghakhanian et al. 2015).This strongly suggests that haplogroup
D does not indicate a separate ancestry for Andamanese populations.
Rather, haplogroup D was part of the
standing variation carried by the OOA expansion, and later
lost from most of the populations except in Andaman and
partially in Japan and Tibet.