Extended Data Fig. 4: Fine-scale genetic structure within Oceania.
From: Indigenous Australian genomes show deep structure and rich novel variation
A. Hierarchical clustering tree (up to 13 clusters) produced from the maximum a posteriori state partitions inferred by fineSTRUCTURE (see Methods). The NCIG + PNG (masked) dataset used for this analysis was reduced to a subset of unadmixed and unrelated samples. Samples are coloured according to their sampling location. Note that the 25 samples from Papua New Guinea are denoted by their sub-sampling locations in this analysis (Bundi, Kundiawa (Kuman), Marawaka, Mendi and Tari). B. Clustering of the NCIG + PNG (masked) dataset into 5 clusters based only on genetic data using fineSTRUCTURE. The dataset used for this analysis was reduced to a subset of unadmixed and unrelated samples. For each individual, the coloured symbol represents the genetic cluster to which the individual is assigned. C. The fineSTRUCTURE coincidence matrix showing the proportion of cluster partitions in which two individuals are grouped in the same cluster during the MCMC. The NCIG + PNG (masked) dataset used for this analysis was reduced to a subset of unadmixed and unrelated samples.
