The frequency in Japanese of genetic variants of 22 proteins: V. Summary and comparison with data on Caucasians from the British Isles

Abstract

The frequencies in Hiroshima and Nagasaki of rare variants (represented in less than 2 % of the individuals surveyed) is summarized for a series of 22 proteins (25 polypeptides). The average number of persons examined for each protein was 3312. There are three pairs of homologous proteins in the series: PGM, and PGM, CA I and CA 11, and HGB A and A,. Only for the first pair is there a significant difference between the two in the total frequency and number of different kinds of variants; it is suggested this may reflect differences in the mutation rates of the corresponding structural genes. For 23 of these polypeptides, comparable data are available for British Caucasians. The average frequency of variants for loci in common in the two series is 2.0/1000 person determinations for Japanese and 1.6/1000 for Caucasoids. At two loci (PGMl and PHI) there were significantly more variants in Japanese than in British; these two loci account for the greater average frequency of variants in Japanese. However, a conservative comparison of number of diflerent variants (electromorphs) encountered, using the 0 statistic of Ewens (1972), yields no significant difference for any of the 22 possible contrasts. The potential usefulness of data of this type in reaching conclusions regarding comparability of mutation rates in two populations is discussed. For the present, the fact that one electromorph may shelter multiple different amino acid substitutions in a protein limits the inferences to be drawn from such contrasts. It is 8 pleasure to acknowledge our indebtedness to Dr Peter Smouse and Dr Warren Ewens for statistical consultations, but responsibility for the tentative interpretations is entirely our own.