Premium
A quantitative review of MHC ‐based mating preference: the role of diversity and dissimilarity
Author(s) -
Kamiya T.,
O'Dwyer K.,
Westerdahl H.,
Senior A.,
Nakagawa S.
Publication year - 2014
Publication title -
molecular ecology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.619
H-Index - 225
eISSN - 1365-294X
pISSN - 0962-1083
DOI - 10.1111/mec.12934
Subject(s) - biology , mate choice , evolutionary biology , diversity (politics) , major histocompatibility complex , sexual selection , preference , mating preferences , mating , phylogenetic tree , selection (genetic algorithm) , range (aeronautics) , taxon , ecology , genetics , gene , statistics , materials science , mathematics , composite material , artificial intelligence , sociology , anthropology , computer science
Sexual selection hypotheses stipulate that the major histocompatibility complex genes ( MHC ) constitute a key molecular underpinning for mate choice in vertebrates. The last four decades saw growing empirical literature on the role of MHC diversity and dissimilarity in mate choice for a wide range of vertebrate animals, but with mixed support for its significance in natural populations. Using formal phylogenetic meta‐analysis and meta‐regression techniques, we quantitatively review the existing literature on MHC ‐dependent mating preferences in nonhuman vertebrates with a focus on the role of MHC diversity and dissimilarity. Overall, we found small, statistically nonsignificant, average effect sizes for both diversity‐ and dissimilarity‐based mate choice ( r = 0.113 and 0.064, respectively). Importantly, however, meta‐regression models revealed statistically significant support regarding female choice for diversity, and choice for dissimilarity (regardless of choosy sex) only when dissimilarity is characterized across multiple loci. Little difference was found among vertebrate taxa; however, the lack of statistical power meant statistically significant effects were limited to some taxa. We found little sign of publication bias; thus, our results are likely to be robust. In light of our quantitative assessment, methodological improvements and fruitful future avenues of research are highlighted.