Date: Wed, 13 May 1998 11:36:42 +0100 (BST)
From: Caroline Osella <Caroline.Osella-AT-durham.ac.uk>
Subject: genetics_of_castes (fwd)


This is the reply to my mail.


Caroline Osella
***************************************************************

---------- Forwarded message ----------
Date: Thu, 7 May 1998 09:46:11 -0600 (MDT)
From: Michael Bamshad <mike-AT-genetics.utah.edu>
To: Caroline.Osella-AT-durham.ac.uk
Cc: mike-AT-thor.genetics.utah.edu, filippo.osella-AT-durham.ac.uk
Subject: genetics_of_castes

Dear Mrs Osella,

Malia forwarded your e-mail message to me and suggested that I contact you
to directly answer your questions about our work with South Indian populations.
One big disadvantage that I have is that I have not read any of the articles
about our work that have been published in England or India, and from what sort
of inquiries I have received, I suspect the information has been widely
disseminated.  Unfortunately, I think the press has promoted a rather preliminary
finding and turned it into fact.

The work that was presented at the AAPA meetings examined the relationship
between caste rank and genetic distances between castes.  If we look at
a genetic system that is transmitted only through maternal lineages (i.e., 
mitochondrial DNA or mtDNA) we find that genetic distance is correlated to
social rank.  Thus, the genetic distance between upper and lower castes is
higher than the genetic distance between upper and middle or middle and lower
castes.  On the other hand, if we look at a genetic system that is transmitted
only through paternal lineages (i.e., Y chromosome polymorphisms) we find
that there is no correlation between genetic distance and social rank.  There
are a few good explanations for these findings, but we have argued (and I
think our data support our conclusions) that this pattern is most likely the
result of female movement (gene flow) between castes--likely from lower to
higher castes, while males remain in the castes in which they were born
(thus the pattern of Y differences is more likely due to genetic drift 
and mutation).

With respect to continental populations, we do find substantial evidence
that upper castes are more closely related to European populations than are
the middle or lower castes.  The finding that the lower castes or a tribal
population are more similar to Africans comes from earlier work, but the
sample sizes and number of polymorphisms assayed was much smaller and the
finding was very, very preliminary.  The press seems to have become aware of
our caste work and since there is an embargo on the release of any
further information until our work is peer-reviewed and published, I fear
the press has dug up the earlier work and is misreprenting the conclusions.
However, we are very actively working to determine whether castes of different
rank are admixed to different degrees with various continental populations. I
think that this is likely to be true, but what exactly are going to be the
patterns of admixture is unclear.  I anticipate that the composition of 
castes is a mix between varying amounts of indigenous genes, admixture from
different continental groups, on top of which are the complex patterns created
by the caste system.  I think that this complexity is one of the reasons that 
South Indian populations have been so recalcitrant to genetic analysis.  
Additionally, our sampling scheme and populations chosen for study were done
so very carefully.  However, it will be necessary to repeat this analysis
in another location, and we will pursue that.

Throughout this study we have been very careful to attempt to place this work
in the appropriate social and political context.  However, much of what has
happened in the press is beyond our control.  More important to us is that 
our study participants and collaborators in India understand the logic behind
this work and our dedication to helping them start to do molecular work in their
labs in India.  We are working with about 10 different Indian anthropologists
spread throughout South India and in additiona to the castes, about 20 different 
tribal populations.  Hopefully, in the next few years the complex relationships 
between these groups will be better understood.

I'd be interested in knowing what is the nature of your work in South India? and 
where in South India do you work?  

If you have any additional questions or comments, please feel free to contact me.

With sincere regards,

Mike Bamshad



     --- from list seminar-12-AT-lists.village.virginia.edu ---