Fig. 3: Historical relationships between Australian and PNG populations.
From: Indigenous Australian genomes show deep structure and rich novel variation
a, Top, shared genetic drift between populations estimated by outgroup F3 statistics of the form F3 (Yoruba; PNG, X), where X is an Australian individual. Higher values indicate greater shared genetic drift with PNG. Individuals are rank-ordered by F3 value within populations, with block jackknife-estimated standard errors shown as vertical bars. The range of F3 values for individuals in the Tiwi and Galiwin’ku population samples is indicated by horizontal shading. Bottom, the proportion of Papuan global ancestry (after masking) estimated by RFMIX for the same individuals. These per-individual metrics include related individuals. Sample sizes: Tiwi n = 48, Galiwin’ku n = 38, Titjikala n = 13, Yarrabah n = 45. b, Z-scores derived from F4 statistics of the form \({F}_{4}^{({\rm{T}})}\)(Asia-Y, Yoruba; Australia-X, Titjikala), where Asia-Y is a Eurasian or Oceanic population sample from SGDP and Australia-X is either the Galiwin’ku or Tiwi Islands sample. Z-score values greater than 3 provide statistically significant evidence that population Asia-Y shares more genetic drift with Tiwi/Galiwin’ku than with Titjikala, and these populations are marked with an asterisk. The per-individual metrics include related individuals.
