Table 2.
Estimates of the mean number of nonsynonymous substitutions, the relative nonsynonymous substitution rate in the PBR, and the selection coefficient (s)
| HLA locus | Lengtha | L S a | L B a | L N a | No. of allele 1b | No. of allele 2c | K B | γ | S | M | s |
|---|---|---|---|---|---|---|---|---|---|---|---|
| A | 1,095 bp | 295 | 123 | 674 | 27d | 9 | 28.9 (26.0) | 7.6 (6.3) | 4,500 (3,000) | 0.04 (0.09) | 2.25 % (1.50 %) |
| B | 1,086 bp | 300 | 122 | 643 | 40d | 19 | 35.9 (36.0) | 9.7 (9.0) | 8,825 (8,200) | 0.01 (0.02) | 4.41 % (4.20 %) |
| C | 1,093 bp | 301 | 125 | 665 | 20d | 20 | 17.3 (15.0) | 4.9 (3.4) | 1,030 (530) | 0.15 (0.29) | 0.52 % (0.26 %) |
| DRB1 | 795 bp | 223 | 53 | 521 | 26 | 25 | 23.2 (25.0) | 10.2 (9.3) | 3,890 (3,900) | 0.01 (0.01) | 1.94 % (1.90 %) |
| DQB1 | 687 bp | 148 | 51 | 347 | 13 | 13 | 12.4 (20.0) | 4.4 (6.0) | 479 (1,700) | 0.14 (0.08) | 0.24 % (0.85 %) |
| DPB1 | 543 bp | 146 | 53 | 344 | 11 | 10 | 11.9 (6.8) | 9.2 (4.3) | 918 (140) | 0.01 (0.08) | 0.46 % (0.07 %) |
| DRB3 | 549 bp | 148 | 54 | 347 | (13e) | − | 5.6 − | 5.4 − | 120 − | 0.04 − | 0.06 % − |
| DRB5 | 549 bp | 148 | 53 | 348 | (5e) | − | 8.0 − | 7.9 − | 360 − | 0.01 − | 0.18 % − |
| DQA1 | 765 bp | 211 | 47 | 504 | 8 | 6 | 5.9 (13.0) | 2.1 (4.5) | 53 (550) | 0.23 (0.14) | 0.03 % (0.28 %) |
| DPA1 | 663 bp | 190 | 42 | 428 | 5 | 3 | 4.8 − | 3.3 − | 54 − | 0.10 − | 0.03 % − |
The numbers of sites of synonymous and nonsynonymous substitutions were estimated using the modified Nei–Gojobori model (R = 1.04 for class I, R = 1.14 for class II). The parameter values in parentheses were estimated on the basis of method II described in Satta (1992). The mutation rate per PBR per generation (u) = 1.7 × 10−6 for class I loci and 7.5 × 10−7 for class II loci; effective population size (N e) = 105 (see Satta et al. 1994)
L S the number of synonymous sites across the entire region, L B the number of nonsynonymous sites at the PBR, L N the number of nonsynonymous sites at the non-PBR
aThe length or the number of sites used in this study (not in the previous study)
bThe number of dominant alleles that have a high frequency (>1 %) throughout human populations worldwide (shown as n a in text)
cThe number of dominant alleles excluding possible recombinants
dThe number of dominant alleles that are detected in >100 chromosomes from >25 human populations
eThe number of alleles not derived from the dominant allele because of lack of information about allele frequencies in the human populations