PDA

View Full Version : Ancestral genome results


Birchy
06-30-2026, 04:00 PM
An ok service for a £14 g25 upload but does anyone know why ANF is heavily inflated in their models? I should only be roughly 36% ANF based on other models like QPADM
148874
148875
148876
148877
148878
148879
148880
Mr. M
06-30-2026, 05:52 PM
G25 is well known for overestimating ANF+EHG in Europeans
celticdragongod
06-30-2026, 11:30 PM
An ok service for a £14 g25 upload but does anyone know why ANF is heavily inflated in their models? I should only be roughly 36% ANF based on other models like QPADM
148874
148875
148876
148877
148878
148879
148880

I think Yamnaya/Steppe is made up of EHG, CHG and EEF so the above model strips out EEF from the Steppe and combines it with rest of the EEF.
celticdragongod
06-30-2026, 11:31 PM
G25 is well known for overestimating ANF+EHG in Europeans

I've never heard this before. Do you have any sources related to this?
Opie
07-01-2026, 12:41 AM
G25 is well known for overestimating ANF+EHG in Europeans

And underestimates Zagros/CHG.
Gannicus
07-01-2026, 02:15 PM
And underestimates Zagros/CHG.

I've seen this too. G25 will not assign GanjDareh if I use it as a source for myself, but qpAdm does.