Statistical studies on protein polymorphism in natural populations. I. Distribution of single locus heterozygosity

Surveying the literature, the frequency distribution of single-locus heterozygosity among protein loci was examined in 95 vertebrate and 34 invertebrate species with the aim of testing the validity of the mutation-drift hypothesis. This distribution did not differ significantly from that expected under the mutation-drift hypothesis for any of the species examined when tested by the Kolmogorov-Smirnov goodness-of-fit statistic. The agreement between the observed interlocus variance of heterozygosity and its theoretical expectation was also satisfactory. There was an indication that variation in the mutation rate among loci inflates the interlocus variance of heterozygosity. The variance of heterozygosity for a homologous locus among different species was also studied. This variance generally agreed with the theoretical value very well, though in some groups of Drosophila species there was a significant discrepancy. The observed relationship between average heterozygosity and the proportion of polymorphic loci was in good agreement with the theoretical relationship. It was concluded that, with respect to the pattern of distribution of heterozygosity, the majority of data on protein polymorphisms are consistent with the mutation-drift hypothesis. After examining alternative possible explanations involving selection, it was concluded that the present data cannot be explained adequately without considering a large effect of random genetic drift, whether there is selection or not.