Socially Structured Populations and Evolution of Recombination under Antagonistic Coevolution

Abstract

Theoretical studies have helped to clarify under which conditions a modifier for recombination rate can spread in a population. In large populations and with no epistasis or linkage, genes that affect recombination rate are selectively neutral (Maynard Smith 1968). With directional selection, evolution toward recombination is possible with sufficiently weak negative epistasis but never under positive epistasis (Barton 1995). Negative epistasis occurs when the fitness of an offspring carrying two beneficial alleles is less than the product of the fitnesses of offspring carrying one allele each. Weak epistasis is required because of the shortterm effect of negative epistasis to reduce the fitness of recombinants, which must be outweighed by the longterm advantage of recombining beneficial alleles onto the same genotype (Barton and Charlesworth 1998). Populations indeed show increased rates of recombination under experimental directional selection (Korol 1999) or as a result of improving stock quality in domestic animals (Burt and Bell 1987). In this view, the recombination modifier hitchhikes on the long-term fitness advantages of good genes. Alternatively, short-term processes could be important too. Novel gene combinations in offspring could have a disproportional advantage in the next or immediately following generations. According to theory, this is possible when the sign of epistasis (positive or negative) changes