Complex signatures of natural selection at the Duffy blood group locus

Abstract

The Duffy blood group locus (FY) has long been considered a likely target of natural selection, because of the extreme pattern of geographic differentiation of its three major alleles (FY*B, FY*A, and FY*O). In the present study, we resequenced the FY region in samples of Hausa from Cameroon (fixed for FY*O), Han Chinese (fixed for FY*A), Italians, and Pakistanis. Our goals were to characterize the signature of directional selection on FY*O in sub-Saharan Africa and to understand the extent to which natural selection has also played a role in the extreme geographic differentiation of the other derived allele at this locus, FY*A. The data from the FY region are compared with the patterns of variation observed at 10 unlinked, putatively neutral loci from the same populations, as well as with theoretical expectations from the neutral-equilibrium model. The FY region in the Hausa shows evidence of directional selection in two independent properties of the data (i.e., level of sequence variation and frequency spectrum), observations that are consistent with the FY*O mutation being the target. The Italian and Chinese FY data show patterns of variation that are very unusual, particularly with regard to frequency spectrum and linkage disequilibrium, but do not fit the predictions of any simple model of selection. These patterns may represent a more complex and previously unrecognized signature of positive selection.

Figure 1

FY-gene region. Numbers (×1,000) below the line refer to the nucleotide position from GenBank accession number AL035403. The shaded vertical bars (numbered above the line) indicate the positions of exons in a putative BL2-like gene 5′ to the FY locus. The inverted triangle indicates the position of the FY*O mutation 5′ to the FY coding region (blackened horizontal bar). Thick lines below the numbered line indicate the regions surveyed for sequence variation in each population. Roman numerals indicate how the sequence data were partitioned into segments for the purpose of data analysis. This partition was arbitrary and simply reflects the boundaries of the segments surveyed in different population samples.

Figure 2

Comparison of F_ST in the FY region and in the 10 locus-pairs regions. Blackened diamonds represent F_ST at various distances from the FY gene. Position 0 represents the location of the FY*A mutation in the top panel and the FY*O mutation in the bottom panel. Regions II and III were pooled and then divided into three regions of ∼2.5 kb, the average size of the locus-pairs regions. The shaded area represents the range of values of F_ST observed, between these populations, in the locus-pairs regions. The dashed line represents the average F_ST in the locus-pairs regions.

Figure 3

Neighbor-joining tree of region III, for the Hausa, Chinese, and Italians. Haplotypes inferred by PHASE were used in the analysis and are represented only once, even if they occurred multiple times in a sample. The FY*O and FY*A sites were omitted from the sequence of the inferred haplotypes.

Figure 4

Sliding window of polymorphism and divergence in the Chinese sample. Regions III–VII are represented. The solid line indicates nucleotide diversity; the dashed line indicates sequence divergence from orangutan. The window size is 1,000 bp.

Figure 5

LD in regions II–VIII of the Chinese sample. |D′| (top) and r² (bottom) were estimated for pairs of sites in diploid data. The shading of squares indicates the value of the statistic: blackened squares = .75–1.0; dark shaded squares = .5–.75; medium shaded squares = .25–.5; and light shaded squares = 0–.25. The numbers indicate the position of the polymorphic site. The position of the FY gene is marked by a bar (left).

Figure 6

Estimated haplotypes for regions II–VIII. The two haplotypes inferred from one diploid genotype are labeled “A” and “B” (e.g., IT1A and IT1B). Haplotypes within a population sample are grouped, to emphasize haplotype structure. In most cases, dark shading indicates a derived allele, and light shading indicates an ancestral allele. An asterisk (*) indicates that the ancestral state is unknown, in which case the color is arbitrary. A hatch (#) indicates that the ancestral state at CpG sites is ambiguous, in which case C was assumed to be ancestral. White gaps indicate that the sites were not scored in the sample.

Figure 6

Estimated haplotypes for regions II–VIII. The two haplotypes inferred from one diploid genotype are labeled “A” and “B” (e.g., IT1A and IT1B). Haplotypes within a population sample are grouped, to emphasize haplotype structure. In most cases, dark shading indicates a derived allele, and light shading indicates an ancestral allele. An asterisk (*) indicates that the ancestral state is unknown, in which case the color is arbitrary. A hatch (#) indicates that the ancestral state at CpG sites is ambiguous, in which case C was assumed to be ancestral. White gaps indicate that the sites were not scored in the sample.