HLA Diversity/Anthropology

Henry Erlich, Chair

The goal of the Diversity/Anthropology Component is to determine HLA class I and class II allele and haplotype frequencies in various human populations. Studies of allelic diversity in different populations can shed light on the evolution of HLA polymorphism as well as on the evolution and migration of human populations. In a clinical context, a knowledge of the allele frequency distributions in various populations is critical to the strategy of establishing and searching bone marrow donor registries as well as in studies of HLA-associated disease susceptibility.

At the ASHI meeting in Atlanta, a group of about 20 scientists interested in the HLA-Diversity/Anthropology Workshop met to discuss a number of issues. Some of the issues discussed were:

1. Resolution of the Typing Method

It was decided unanimously that the typing for samples to be included in the Diversity/Anthropology Component should be carried out in high resolution. Because the samples to be included in the bone marrow donor registry component will probably be typed at intermediate resolution, the recommendation was that, for inclusion into the Diversity/Anthropology component, a subset of the registry samples would be retyped at a higher level of resolution.

2. Standardization of a Typing Method

Because this issue was not yet resolved within the larger framework of the Workshop, the group thought this issue should be deferred until it is addressed and resolved for the whole Workshop. Whatever typing method(s) were used, it was recommended that a blind QC/proficiency panel be included to provide a "filter" to ensure high quality data.

3. Selection of Populations

This issue has an inevitable aspect of "trade-off" in that the "ideal" populations (in terms of anthropological definition) may be very difficult to sample, for a variety of reasons, and samples that are more readily available, but may not be anthropologically "ideal" are more likely to be included into a typing program. It was suggested that those population samples that were typed at a low-intermediate level with the class I ARMS method for the 12th Workshop could be typed at high resolution for the 13th Workshop. Julia Bodmer has made the names and addresses of the contributing labs available and they will be contacted to see if there is interest in participating in the high level typing.

The definition of new populations to be typed will be an ongoing issue for discussion, and suggestions are very welcome, both in terms of what populations might be most informative, from an anthropological genetics perspective, and what populations might be accessible. The issue of "isolated indigenous" vs. so-called "census" populations was discussed as well as the problem of admixture for certain kinds of analytic approaches. There may be labs and investigators outside the HLA field who would like to collaborate with the Workshop in the study of HLA diversity of their samples. Also, as noted above, the group thought that a subset of the samples from the bone marrow donor registry component, should be typed at high resolution to provide data from a more general and less geographically defined populations sample.

4. Ethical/Legal Issues and Informed consent

It was reminded that those dimensions of the work were of primary importance and that the knowledge and respect of regulations for those aspects in the country/region where populations were sampled were an absolute requisite.

5. Population Size and Structure

There was a consensus that the size of the populations sampled and typed should be sufficient to provide reliable estimates of allele frequencies. Populations for which the number of samples is substantially less tan 50 will not be very informative. IF at all possible, it would be desirable for the population to be sufficiently large to be able to estimate haplotype frequencies as well although this goal, other than DR-DQ haplotypes, may be difficult to achieve, given the size of some of the populations samples. Information about inbreeding pedigrees and population structure would also be valuable, whenever possible.

6. Markers

At a minimum, data on HLA-A, -B, -C, as well as the DRB loci DQB1, DQA1, DPB1, and DPA1 would be included. Linkage disequilibrium patterns for these HLA loci would be calculated, when possible. The possibility of analyzing microsatellite markers in the MHC was also raised as was the analysis of intron sequences. Suggestions are welcome.

7. Analytic Approaches

Allele frequencies will be calculated and haplotype frequencies, whenever possible, will be estimated. The Ewens-Watterson F-statistic analysis of selection will be applied to the allele frequency distributions. Phylogenetic inferences will be carried out using tree-building algorithms as well as various statistical approaches (e.g., distance from Centroid). Suggestions or other approaches are welcome.

8. Organizing Committee

The current committee members include several people with extensive experience in previous HLA Diversity/Anthropology Components. Dr. Erlich has invited two distinguished anthropologic geneticists from outside the HLA field, Mark Stoneking and Michael Crawford, to join the committee. The current committee members are listed below. Please contact the committee members or Dr. Erlich with any suggestions about this component.

P. Beatty (USA)
J. Bodmer (UK)
A. Cambon-Thomsen (France)
M. Crawford (USA)
M. Fernandez-Vina (USA)
X. Gao (Australia)
T. Gojobori (Japan)
W. Klitz (USA)
S. Mack (USA)
E. Milford (USA)
T. Mori (USA)
M. Stoneking (USA)

Human Diversity Survey