Masatoshi Nei

Japanese American geneticist (1931–2023)

Masatoshi Nei (根井正利) was a Japanese born American evolutionary biologist.

Masatoshi Nei was born in 1931 in Miyazaki Prefecture, on Kyūshū Island, Japan. He received a Bachelor of Science degree from the University of Miyazaki in 1953, and published his first article, on the mathematics of plant breeding, that same year. In 1959, he completed his doctoral degree at Kyoto University on quantitative genetics for crop improvement.

Nei was the first to show mathematically that, in the presence of gene interaction, natural selection always tends to enhance the linkage intensity between genetic loci or maintain the same linkage relationship. He then observed that the average recombination value per genome is generally lower in higher organisms than in lower organisms and attributed this observation to his theory of linkage modification. Recent molecular data indicate that many sets of interacting genes such as the Hox genes, immunoglobulin genes, and histone genes have often existed as gene clusters for a long evolutionary time. This observation can also be explained by his theory of linkage modification. He also showed that, unlike R. A. Fisher's argument, deleterious mutations can accumulate rather quickly on the Y chromosome or duplicate genes in finite populations. In 1969, considering the rates of amino acid substitution, gene duplication, and gene inactivation, he predicted that higher organisms contain a large number of duplicate genes and nonfunctional genes (now called pseudogenes). This prediction was shown to be correct when many multigene families and pseudogenes were discovered in the 1980s and 1990s. His notable contribution in the early 1970s is the proposal of a new measure of genetic distance (Nei's distance) between populations and its use for studying evolutionary relationships of populations or closely related species. He later developed another distance measure called DA, which is appropriate for finding the topology of a phylogenetic tree of populations. He also developed statistics of measuring the extent of population differentiation for any types of mating system using the GST measure. In 1975, he and collaborators presented a mathematical formulation of population bottleneck effects and clarified the genetic meaning of bottleneck effects. In 1979, he proposed a statistical measure called nucleotide diversity, which is now widely used for measuring the extent of nucleotide polymorphism. He also developed several different models of speciation and concluded that the reproductive isolation between species occurs as a passive process of accumulation of interspecific incompatibility mutations

In the early 1960s and 1970s, there was a great controversy over the mechanism of protein evolution and the maintenance of protein polymorphism. Nei and his collaborators developed various statistical methods for testing the neutral theory of molecular evolution using polymorphism data. Their analysis of the allele frequency distribution, the relationship between average heterozygosity and protein divergence between species, etc., showed that a large portion of protein polymorphism can be explained by neutral theory. The only exception was the major histocompatibility complex (MHC) loci, which show an extraordinarily high degree of polymorphism. For these reasons, he accepted the neutral theory of evolution.

Using his genetic distance theory, he and A. K. Roychoudhury showed that the genetic variation between Europeans, Asians, and Africans is only about 11 percent of the total genetic variation of the human population. They then estimated that Europeans and Asians diverged about 55,000 years ago and these two populations diverged from Africans about 115,000 years ago. This conclusion was supported by many later studies using larger numbers of genes and populations, and the estimates appear to be still roughly correct. This finding could be considered an early indication of the out of Africa theory of human origins.

Around 1980, Nei and his students initiated a study of inference of phylogenetic trees based on distance data. In 1985, they developed a statistical method for testing the accuracy of a phylogenetic tree by examining the statistical significance of interior branch lengths. They then developed the neighbor joining and minimum evolution methods of tree inference. They also developed statistical methods for estimating evolutionary times from molecular phylogenies. In collaboration with Sudhir Kumar and Koichiro Tamura, he developed a widely used computer program package for phylogenetic analysis called MEGA.

Nei's group invented a simple statistical method for detecting positive Darwinian selection by comparing the numbers of synonymous nucleotide substitutions and nonsynonymous nucleotide substitutions. Applying this method, they showed that the exceptionally high degree of sequence polymorphism at MHC loci is caused by overdominant selection. Although various statistical methods for this test have been developed later, their original methods are still widely used.