>>213659359
The paper explicitly said ORCHESTRA is a poor tool for admixture events predating the late medieval period and has a hard time differentiating between closely related ancestries, and also assigns ancestry groups that aren't available to the "closest available" modern population
Ashkenazi Jews being 20% Slavic makes their total ancestry similar to North/Central Italians which is what this paper uses as a reference population so it sucks up this ancestry
This is corroborated by the supplemental in the study which has Italian Jews (Roman Jews) as 40% Levantine and 50% Ashkenazi which makes no sense from historical or genetic perspective
Also all Jewish populations across the Mediterranean have Southern European ancestry which this doesn't pick up for the same reasons