Making Vahaduo-like models based on SNP-level data?

[Komintasavalta]

from which study is the Lazaridis dataset exactly? there are many Lazaridis studies.

The datasets from Lazaridis et al. 2014 and 2016 can be downloaded from here: https://reich.hms.harvard.edu/datasets. 1240K+HO has a sample with the same ID for all 2068 samples in the Lazaridis 2016 dataset, and for 1945 out of 1963 of the samples in the Lazaridis 2014 dataset (but the remaining 18 samples consist of some standard ancient samples, a chimp reference genome, and 3 Australian samples that have a different ID). Almost all of the samples in Lazaridis 2016 are also in Lazaridis 2014.

$ curl https://reichdata.hms.harvard.edu/pub/datasets/amh_repo/curated_releases/V44/V44.3/SHARE/public.dir/v44.3_HO_public.anno -Lso ho.anno
$ wc -l NearEastPublic/HumanOriginsPublic2068.ind # Lazaridis 2016
2068
$ awk '{print$1}' NearEastPublic/HumanOriginsPublic2068.ind|awk -F\\t 'NR==FNR{a[$0];next}$2 in a{print$4}' - ho.anno|sort|uniq -c|sort
      5 PickrellNatureCommunications2012
    127 LazaridisNature2016
    881 PattersonGenetics2012
   1055 LazaridisNature2014
$ wc -l EuropeFullyPublic/data.ind # Lazaridis 2014
1963
$ awk '{print$1}' EuropeFullyPublic/data.ind|awk -F\\t 'NR==FNR{a[$0];next}$2 in a{print$4}' - ho.anno|sort|uniq -c|sort
      1 Genome
      5 PickrellNatureCommunications2012
    887 PattersonGenetics2012
   1052 LazaridisNature2014
$ awk 'NR==FNR{a[$1];next}!($1 in a)' EuropeFullyPublic/data.ind NearEastPublic/HumanOriginsPublic2068.ind|wc -l
130