As mentioned earlier, I removed San and Pygmy groups from my reference datasets.
For the admixture runs on Reference Dataset I, the only major changes are for K=2 ancestral components where most European, Middle Eastern and South/Central Asian groups increase their African component. The changes for K=3,4,5 were minor as shown by these statistics:
| K | Median Abs | Maximum Abs |
|---|---|---|
| 3 | 0.01% | 0.22% |
| 4 | 0.02% | 0.26% |
| 5 | 0.02% | 0.71% |
I have updated the spreadsheet and the plots in the original post.
Looking at the changes in the admixture results I already posted for Harappa Project participants HRP0001 to HRP0010, there is major change for K=2. The African compoent (C1/red) increased by a lot among all project participants. This seems to be due to the African component best representing West Africans now instead of Pygmies as it did before.
For K=3,4,5, the changes are very minor. Let's look at the absolute value of the changes in the percentages of ancestral components for the ten project participants.
| K | Median Abs | Maximum Abs |
|---|---|---|
| 3 | 0.05% | 0.19% |
| 4 | 0.05% | 0.22% |
| 5 | 0.09% | 0.60% |
I have updated the spreadsheets and the charts in the original post.
Comments are closed.