Insights into the genetic architecture of morphological and sexually selected traits in two passerine bird species

Abstract

Knowledge about the underlying genetic architecture of phenotypic traits is needed to understand and predict their evolutionary dynamics. The number of causal loci underlying most phenotypic traits, the magnitude of their effects and their location in the genome is still largely unknown. Here we use genome-wide SNP data from two large-scale datasets on house sparrows and collared flycatchers to examine the genetic architecture of morphological (tarsus length, wing length, body mass, bill depth and bill length) and sexually selected traits (total badge size and visible badge size in house sparrows and white wing patches in collared flycatchers). Heritabilities were estimated using kinship coefficients calculated from the SNP data. The proportion of variance captured by the SNPs (SNP-based heritability) was lower in house sparrows compared to collared flycatchers, as expected given marker density (6,348 SNPs in house sparrows versus 38,689 SNPs in collared flycatchers). Chromosome partitioning analyses demonstrated that the proportion of variance explained by each chromosome was significantly positively related to the chromosome size for most traits. Finally, we only found two genome-wide significant associations with very small effect sizes; one SNP on chromosome 20 associated with bill length in house sparrows and one SNP on chromosome 4 for tarsus length in collared flycatchers. Although we cannot exclude the possibility of undetected large-effect QTL, our results support the hypothesis that variation in morphological and sexually selected traits is typically determined by many loci with small effect.