Anthropology, Genetic Diversity, and Ethics 
A workshop at the Center for Twentieth Century Studies 
University of Wisconsin-Milwaukee  
Mike Bamshad*
[Participant Information]

I think Eric and Frank have made a number of important and provocative comments.  What I'm going to present to you today is different; it's work in progress, it's work in which we are doing genetic analysis on socially defined groups.  We've tried to address some of the issues which have already been raised, but we've probably not addressed all of them.  We probably could have addressed some of them better, but we think that we've made efforts in many ways to stimulate discussion and to actually bring into the field some of the concerns that people have raised in forums such as this.  So for the next 10 to 15 minutes, I want to introduce some of the strategies and objectives of an ongoing research program to better understand the genetic evolution of South Indian populations.  It's in the context of this project that Lynn Jorde and I proposed to develop and optimize technologies to collect and analyze DNA from aboriginal populations in developing countries.  It was hoped that the experiences gained in this effort would help guide the sampling of the worldwide populations. 

Specifically today, I want to make four points.  I want to explain to you the rationale for studying these Indian populations, I want to identify some of the groups with which we were working, introduce some of our goals, and illustrate some of the preliminary work completed toward these goals.  Tomorrow, Lynn Jorde is going to discuss some of the conclusions drawn from these studies, and J.M. Naidu, one of our Indian collaborators, is going to talk about the reactions to our work from various groups in India. 

The people of India have at least three putative origins: first, Paleolithic settlement of proto-Asians, Africans, and Australoids; second, migration of proto-Dravidians from central Asia, approximately 10,000 years ago; and third, migration of Indo-European-speaking populations from western Asia into the northern part of the Indian subcontinent approximately two to four thousand years before present.  It was this invasion that putatively established the Hindu caste system. This is the history of India as professed by largely European and American historians and linguists.  However some elements of these stories that have been undermined as of late because of the discordance with the archaeological record.  Thus we thought this was an attractive area in which we could apply some molecular tools to further clarify the history of human evolution in India. 

The Hindu caste system is stratified into four varna: the Sudras, the Vaisyas, the Ksatriyas, and the Brahmin.  A fifth varna, the Panchama, was added at a later date.  This stratification is such that the Panchama are considered the lowest castes (i.e., lowest status) while the Brahmin are considered the uppermost caste (i.e., highest status).  Religious rights, material resources and education, and occupational specialization were, and remain, largely determined by caste status.  This hierarchy is maintained by law, ritual and religious philosophy. 

Each caste represents an endogamous.  Marriages occur primarily between individuals of equal status.  Men from a higher varna may occasionally marry women of a lower varna, and this is an accepted marriage.  Marriage of men from a lower varna to women from a higher varna is strictly discouraged.  This sets up a specific model of gene flow through the caste system, such that women are much more likely to move from a lower caste to a higher caste, with more gene flow occurring between castes which are adjacent to one another.  However, men are relatively fixed within each of their caste levels.  This model has been of longstanding study by many social scientists, and given that Hindus comprise approximately 80 percent of the population of India, or nearly one billion people, it was also of interest to population geneticists like ourselves.  Thus we set out to test whether this model was consistent with empirical molecular data. 

This photograph illustrates one of the populations with which we've been working.  This is the village of the Yadava community, an agricultural community.  All of the populations we studied were sampled from a single state and a single district within a single state in south India.  These men are members of another caste community, the Jalari.  They're a fisherman caste, also a middle-level caste community.  To draw comparisons to non-caste communities, we've also collaborated with approximately ten physical anthropologists from various universities in different states of India, to collect genetic data from tribal populations. To date we've collected phenotypic material and genetic material from fourteen different tribal populations distributed across south India. 

One of the most consistent criticisms of investigators trying to work with local investigators in developing countries to collect and analyze data from aboriginal populations has been the lack of continued support for local investigators to pursue their own academic interests.  So three goals that we established in this project include the establishment of local laboratories for our collaborators, the transfer of a variety of technologies, and the development of methods for the field analysis of molecular markers.  Achievement of these goals would provide our collaborators in these developing countries with the capability to use modern molecular tools to answer the questions that are of greatest interest to them.  In establishing a local laboratory, we needed -- to provide them with equipment, the resources, ongoing support, and technical assistance.  Additionally, we wanted to make certain that they were able to do DNA extractions in the field, so that they could continue to work with us, and collaborate with each other. 

One of the biggest challenges in setting up a local laboratory was ensuring that they had access to water, electricity, and a waste disposal facility.  Additionally, we had to be sensitive to the types of materials that were locally accessible to them, and the disposal of those materials.  All the techniques that we utilize for testing molecular markers in our laboratory had to be adapted to be used at universities in south India, including using chemicals that were not hazardous and easily disposed.  Additionally, we had to find local sources or substitutes for many of the materials that we use in our laboratory and that were not locally available in India. Many of the analyses that we do in our laboratory use a variety of probes, some of which are hazardous, others of which are hazardous and radioactive.  Thus, we had to develop methods that could be used in their laboratory with non-hazardous probes.  This is a photograph of the laboratory as it stands as of last year, when I was last in India.  It's divided into two sections: this is a dry lab area, wet lab area.  We've created a small conference room for them, and we've included a library for students and investigators.  We've also installed what was at that time a fast computer. 

One of the persistent problems has been a lack of access to a consistent power supply, so this computer actually operates on a battery for eight hours of each day.  Unfortunately, brownouts and blackouts are common in India, and they occur typically during the middle of the day, which, as you'll see in a moment, is also a problem when you're trying to do bench experiments using PCR machines and centrifuges, et cetera. This is a photograph that illustrates the wet laboratory.  They've been equipped with centrifuges, vortex machines, pipetters, everything that we use in our laboratory to answer questions of anthropological interest.  Additionally, they now have a  PCR machine.  This is the first PCR machine at the university.  It's become not only a resource for members of the Division of Anthropology, but for many different departments at Andhra University. This PCR machine also runs on a battery. 

One of the most controversial points about working with aboriginal populations has been the collection of whole blood for transformation (i.e., the creation of cell lines).  Thus, one of the things that we wanted to do was to identify culturally acceptable, minimally invasive collection techniques.  For many other reasons too, it's often very challenging to do venipuncture in the field, and collecting large amounts of blood by venipuncture is often seen as quite different and more invasive than collecting blood, for example, by finger prick, where you simply use sharp objects to make a prick in the end of the finger and collect about half a milliliter of blood; but from this half mil of blood, we can collect 10 to 15 micrograms of DNA, which is more than sufficient for our analysis.  Moreover, by modifying the techniques we commonly use and minimizing the amount of DNA required, 10 to 15 micrograms is enough DNA to do 500 to 1000 typical PCR reactions.  PCR reactions are one of the first steps in assaying markers that are used in these genetic analyses.  We've also now been able to combine the testing of multiple markers in a single reaction.  These modifications have circumvented the need to collect blood by venipuncture, as well as the establishment of cell lines.  Additionally, we tested a variety of alternative methods of collecting DNA.  The one that seems to be the most appropriate was the use of a plucked hair. 

We also wanted to optimize the utilization of collected material.  Most of the genetic data that have been generated to date on the Indian populations are variants of red blood cell enzymes, and the serum protein polymorphisms.  Thus, if we wanted to compare our data to the data sets that have been created over the last 20 or 30 years, as well as the data sets that commonly continue to be created, we had to find a way to not destroy some of the cellular material, as well as the serum that was collected when we extracted DNA.  We've done so, and  all this material remains at the local universities, where Indian investigators can pursue the typing of different markers for their own projects, and integrate it with the work that we were doing.  We wanted to explore the use of archival material at many of these universities.  Anthropologists have been collecting samples for tens of years, and have thousands of archived serum samples that they can work with, so we wanted to develop a method that could be used to extract DNA from those archived samples. 
This photograph illustrates one of our collection teams.  We typically recruited assistants from graduate students and faculty at the various universities.  We also tried to incorporate the use of a local Indian physician.  These groups were necessary, because we were collecting genetic material, and a variety of different phenotypic measurements, dermatoglyphics, morphometrics. 

Lastly, to provide ongoing support for the laboratory and to ensure that students and faculty would continue to be trained, we established a training program for students and faculty.  This consists of local workshops, of which we've had two now, training them in different molecular methods and different statistical methods that are used to analyze molecular data.  We had faculty members from Andhra University visiting the University of Utah, working at our laboratory, learning things that they subsequently take back to their university and train people from various departments.  Additionally, we've established not only a major lab now at Andhra University, but a number of satellite labs throughout south India, all within departments of anthropology.  We have encouraged our Indian collaborator to seek independent, extramural funding for their research interests, and a number of them have research proposals now pending with the Indian government. 

Tomorrow, J.M. Naidu is going to discuss the reactions, to the results of our work, which have been widely published in India.  But in general, we have had enthusiastic support from local scientists and students.  We've met with approval, albeit cautious approval, from local and national government agencies.  The feedback from most of the participants has also been very positive.  In summary, we've made rapid progress toward completing the objectives of our initial project, the genetic analysis of south Indian populations and done it in a  way that it can be viewed as a model for U.S. scientists collaborating with scientists in developing countries. 

*This talk has been edited for web publishing by the author.