“…Duplicate reads were marked using picard-tools MarkDuplicates (broadinstitute.github.io/picard). We further downloaded 2, 359 individuals sequenced with 1240k target enrichment capture sequencing (Fer-nandes et al, 2020; Fowler et al, 2022; Fu et al, 2016; Harney et al, 2021, 2022; Kennett et al, 2022; Lazaridis et al, 2022, 2016, 2017; Lipson et al, 2017, 2022; Liu et al, 2022b; Mathieson et al, 2015, 2018; Narasimhan et al, 2019; Novak et al, 2021; Olalde et al, 2018, 2019; Patterson et al, 2022; Prendergast et al, 2019; Reitsema et al, 2022; Rivollat et al, 2020; Sirak et al, 2021; Tiesler et al, 2022; Villalba-Mouco et al, 2021) directly in BAM-format. For both data sets, we used the task in (Link et al, 2017) to count the number of reads that aligned to each chromosome, ignoring duplicates and reads with a mapping quality below 30.…”