2.4 Estimation of ploidy
We estimated the ploidy levels of Androsace samples using nQuire (Weiß et al. , 2018) following the approach by Viruel et al. (2019). To prepare the reference file for nQuire, we extracted the longest exon recovered per gene using bioawk (available at https://github.com/lh3/bioawk) and excluded any genes that received paralog warnings from the initial reference. We then evaluated ploidy by analyzing the delta log-likelihood (Δ logL values produced by nQuire across three models —diploid, triploid, and tetraploid— to identify the best-supported ploidy level.