Fig. 3

Test for neutrality using coalescent simulations. Position in a–e represents ~100 kbp alignment of assembled MiSeq reads for 12 females (n_black = 6, n_red = 6) as in Fig. 2. a–d Shown are sliding-window comparisons of the observed polymorphism at the Red locus with the polymorphism simulated under a neutral model that incorporated the sampling scheme (six alleles for each of the red and black morphs), demography, recombination, and a derived allele frequency (DAF) of 0.144 at the focal site (see Supplementary Methods). Solid lines represent the observed data (cf. Fig. 2) and the grey-shaded regions represent the space encompassed by the 95% confidence intervals derived from the simulations. a Nucleotide diversity for both morphs’ alleles combined. b d_XY between the two morphs’ alleles. c Tajima’s D for both morphs’ alleles combined. d F_ST between the two morphs’ alleles across the Red locus. e A sliding-window implementation of the modified HKA test, taking into account the observed sampling scheme, allele frequencies, mutation rate and recombination rate. One minus the P-value is plotted for each window for DAF of 0.144 (yellow lines) and 0.856 (blue lines), using the chi-squared distribution (dashed lines) and simulations (solid lines). The red horizontal line indicates a P-value threshold of 0.05. The shaded grey bar indicates an insertion of a putative LTR retrotransposon in the Gouldian finch that is not found in zebra finch