Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
Comparative Study
, 12 (4), 602-12

Patterns of Human Diversity, Within and Among Continents, Inferred From Biallelic DNA Polymorphisms

Affiliations
Comparative Study

Patterns of Human Diversity, Within and Among Continents, Inferred From Biallelic DNA Polymorphisms

Chiara Romualdi et al. Genome Res.

Abstract

Previous studies have reported that about 85% of human diversity at Short Tandem Repeat (STR) and Restriction Fragment Length Polymorphism (RFLP) autosomal loci is due to differences between individuals of the same population, whereas differences among continental groups account for only 10% of the overall genetic variance. These findings conflict with popular notions of distinct and relatively homogeneous human races, and may also call into question the apparent usefulness of ethnic classification in, for example, medical diagnostics. Here, we present new data on 21 Alu insertions in 32 populations. We analyze these data along with three other large, globally dispersed data sets consisting of apparently neutral biallelic nuclear markers, as well as with a beta-globin data set possibly subject to selection. We confirm the previous results for the autosomal data, and find a higher diversity among continents for Y-chromosome loci. We also extend the analyses to address two questions: (1) whether differences between continental groups, although small, are nevertheless large enough to confidently assign individuals to their continent on the basis of their genotypes; (2) whether the observed genotypes naturally cluster into continental or population groups when the sample source location is ignored. Using a range of statistical methods, we show that classification errors are at best around 30% for autosomal biallelic polymorphisms and 27% for the Y chromosome. Two data sets suggest the existence of three and four major groups of genotypes worldwide, respectively, and the two groupings are inconsistent. These results suggest that, at random biallelic loci, there is little evidence, if any, of a clear subdivision of humans into biologically defined groups.

Figures

Figure 1
Figure 1
Relationships between the misclassification rate of the discriminant analysis (Y axis) and the number of loci used for the discrimination (X axis), using two different methods and the Alu21 data set. The graphs represent averages over 500 independent replicates. (Broken line) 1NN, (solid line) RM with Bayesian approach. The loci were added sequentially by Montecarlo randomization, so that their order is not expected to affect the misclassification rate.
Figure 2
Figure 2
Geographical distribution of the five data sets.

Similar articles

See all similar articles

Cited by 34 PubMed Central articles

See all "Cited by" articles

Publication types

LinkOut - more resources

Feedback