## Additional file 3: Figure S2. of Sex-linked genomic variation and its relationship to avian plumage dichromatism and sexual selection

2015-09-16T05:00:00Z (GMT) by
Estimates of genetic diversity ratio ( θ Z / θ A $${\theta}_Z/{\theta}_A$$ ; A) and R Z:A (i.e., corrected for mutation-rate biases; B) across different criterions of data filtering for matched pairs of sexually dichromatic (red) and monochromatic (black) bird species. Lines connected estimates from the same sampled individual. Species pairs are arranged such that the phylogenetic distance to Zebra Finch increases from left to right. Seven different filtering criterions were used. In the first round, all mapped RAD loci with coverage higher than one and lower than the individual cutoff (i.e., mean coverage plus two times standard deviation) were used, and the second round excluded RAD loci with either more than 5 % variable sites, or more than 4 variable sites segregated in 10bp fragment, or more than 20 % sequence divergence from the reference genome. In additional to the second round filtering, we applied another six filters: only including genomic regions shared between the monochromatic and dichromatic species in species pairs; only using mapped RAD loci with sequence coverage ≥5; only using mapped RAD loci with ≤5 %, ≤10 %, ≤15 % sequence divergence from zebra finch; excluding loci from micro-chromosomes (i.e., only have autosomal loci from Chromosome 1-10). For plotting, estimates higher than 1.5 were not shown (marked as x; multiple occurrences for woodpeckers and sapsuckers with stringent divergence filters). Dichromatic species consistently have higher estimates compared to their paring monochromatic species- the highest p values from Wilcoxon signed-rank test on species means were 0.04 ( θ Z / θ A $${\theta}_Z/{\theta}_A$$ ) and 0.02 (R Z:A ) with the first round estimates. (PDF 481 kb)