Reference 3 Admixture K=12

Continuing with the admixture analysis with our new reference 3 dataset.

Here's the results spreadsheet for K=12.

You can click on the legend to the right of the bar chart to sort by different ancestral components.

Of course, the K=11 Onge component was too good to last. Onge are too different from the other populations, so of course they get their isolated component.

Fst divergences between estimated populations for K=12 in the form of an MDS plot.

And the numbers:
C1 C2 C3 C4 C5 C6 C7 C8 C9 C10 C11
C2 0.089
C3 0.093 0.133
C4 0.172 0.211 0.189
C5 0.103 0.080 0.155 0.234
C6 0.094 0.055 0.140 0.218 0.056
C7 0.113 0.143 0.068 0.213 0.169 0.147
C8 0.179 0.219 0.204 0.280 0.237 0.225 0.228
C9 0.177 0.182 0.214 0.285 0.181 0.187 0.232 0.283
C10 0.164 0.178 0.139 0.276 0.214 0.180 0.143 0.290 0.280
C11 0.151 0.150 0.190 0.260 0.150 0.154 0.207 0.262 0.059 0.255
C12 0.256 0.260 0.295 0.373 0.261 0.265 0.314 0.367 0.116 0.364 0.131

3 Comments.

  1. Zack,

    I am not comprehending the way Admixture works. I would appreciate if you could explain briefly how the Kalash component knocked out the Onge one (the Austrics seem to be an exception.).

    Thanks.

    • Kalash component didn't knock out Onge.

      Basically you tell Admixture about the number of ancestral components, K, and it tries to compute those by maximizing loglikelihood based on allele frequencies.

      I expected the Onge component to reduce to just the Onge and a bit of the Great Andamanese since Onge are only distantly related to ASI.

  2. Reference 3 Admixture K=13 | Harappa Ancestry Project - pingback on April 22, 2011 at 10:39 pm

Trackbacks and Pingbacks: