@bumblenick
accounting for differential drift between populations.
Currently available programs such as ADMIXTURE and STRUCTURE are sensitive to differential drift so I wonder if perhaps something using formal statistics like QpAdm(or perhaps that new method) but at the same being able to create genetic clusters like ADMIXTURE is possible. Or something being able to test genetic drift on separate model-based genetic components detected in a given sample, not the whole genome like current programs are doing. I think that would be particularly useful for Paleolithic specimens who according to ADMIXTURE were extremely "mixed". Is that because they weren't fully genetically differentiated yet? Which may complicate correctly detecting their relation to even Mesolithic samples such as WHGs let alone younger populations.
@bumblenick accounting for differential drift between populations. Currently available programs such as ADMIXTURE and STRUCTURE are sensitive to differential drift so I wonder if perhaps something using formal statistics like QpAdm(or perhaps that new method) but at the same being able to create genetic clusters like ADMIXTURE is possible. Or something being able to test genetic drift on separate model-based genetic components detected in a given sample, not the whole genome like current programs are doing. I think that would be particularly useful for Paleolithic specimens who according to ADMIXTURE were extremely "mixed". Is that because they weren't fully genetically differentiated yet? Which may complicate correctly detecting their relation to even Mesolithic samples such as WHGs let alone younger populations.
https://genetiker.wordpress.com/