摘要
Domestication is a selection process that genetically modifies species to meet human needs. A most intriguing feature of domestication is the extreme phenotypic diversification among breeds. What could be the ultimate source of such genetic variations？ Another notable outcome of artificial selection is the reduction In the fitness of domesticated species when they live In the wild without human assistance. The complete sequences of the two subspecies of rice cultlvars provide an opportunity to address these questions. Between the two subspecies, we found much higher rates of non-synonymous （N） than synonymous （S） substitutions and the N/S ratios are higher between cultivars than between wild species. Most interestingly, substitutions of highly dissimilar amino acids that are deleterious and uncommon between natural species are disproportionately common between the two subspecies of rice. We suggest strong selection in the absence of effective recombination may be the driving force, which we called the domestication-associated Hill-Robertson effect. These hitchhiking mutations may contribute to some fitness reduction in cultivars. Comparisons of the two genomes also reveal the existence of highly divergent regions in the genomes. Haplotypes in these regions often form highly polymorphic linkage blocks that are much older than speciation between wild species. Genes from such regions could contribute to the differences between indica and japonica and are likely to be involved in the diversifying selection under domestication. Their existence suggests that the amount of genetic variation within the single progenitor species Oryza rufipogon may be insufficient to account for the variation among rice cultivars, which may come from a more inclusive gene pool comprising most of the A-genome wild species. Genes from the highly polymorphic regions also provide strong support for the Independent domestication of the two subspecies. The genomic variation in rice has revealing implications for studying the genetic b
Domestication is a selection process that genetically modifies species to meet human needs. A most intriguing feature of domestication is the extreme phenotypic diversification among breeds. What could be the ultimate source of such genetic variations？ Another notable outcome of artificial selection is the reduction In the fitness of domesticated species when they live In the wild without human assistance. The complete sequences of the two subspecies of rice cultlvars provide an opportunity to address these questions. Between the two subspecies, we found much higher rates of non-synonymous （N） than synonymous （S） substitutions and the N/S ratios are higher between cultivars than between wild species. Most interestingly, substitutions of highly dissimilar amino acids that are deleterious and uncommon between natural species are disproportionately common between the two subspecies of rice. We suggest strong selection in the absence of effective recombination may be the driving force, which we called the domestication-associated Hill-Robertson effect. These hitchhiking mutations may contribute to some fitness reduction in cultivars. Comparisons of the two genomes also reveal the existence of highly divergent regions in the genomes. Haplotypes in these regions often form highly polymorphic linkage blocks that are much older than speciation between wild species. Genes from such regions could contribute to the differences between indica and japonica and are likely to be involved in the diversifying selection under domestication. Their existence suggests that the amount of genetic variation within the single progenitor species Oryza rufipogon may be insufficient to account for the variation among rice cultivars, which may come from a more inclusive gene pool comprising most of the A-genome wild species. Genes from the highly polymorphic regions also provide strong support for the Independent domestication of the two subspecies. The genomic variation in rice has revealing implications for studying the genetic b
基金
Supported by the National Natural Science Foundation of China （30500049, 30470119 and 30300033）. Tian Tang is supported by International Foundation for Science （IFS）. Publication of this paper is supported by the National Natural Science Foundation of China （30624808）.Acknowledgements We thank Chung-1 Wu at the University of Chicago for many thoughtful comments.
