Additional file 7 of Facilitating population genomics of non-model organisms through optimized experimental design for reduced representation sequencing

• Main Authors: Henrik Christiansen (5717978), Franz M. Heindler (5717981), Bart Hellemans (3246426), Quentin Jossart (532157), Francesca Pasotti (5665906), Henri Robert (6970418), Marie Verheye (11309976), Bruno Danis (811532), Marc Kochzius (242381), Frederik Leliaert (186222), Camille Moreau (811531), Tasnim Patel (11309979), Anton P. Van de Putte (11309982), Ann Vanreusel (220781), Filip A. M. Volckaert (8140614), Isa Schön (153646)
• Format: Other Non-Article Part of Journal/Newspaper
• Language: unknown
• Published: 2021
• Subjects: Genetics; Biodiversity; Genome scan; Genotyping by sequencing; In silico digestion; RADseq; Southern Ocean
• Online Access: https://doi.org/10.6084/m9.figshare.16305513.v1

Additional file 7. Results from parameter optimization for de novo assembly and genotyping. Eight parameter optimization series were conducted following Rochette & Catchen (2017) [45] to identify optimal parameters to genotype reduced representation sequencing (RRS) data with Stacks v2.4 (Rochette et al. 2019) [21]; one test series for each species/species complex. The Stacks parameter m was kept constant (m = 3), while parameters M and n were varied together from 1 to 9. Subsequently, only loci present in 80% of the samples were retained and for each M = n parameter the number of loci and polymorphic loci was plotted, as well as the proportion of these loci containing 0 to 10 or > 10 SNPs. In ostracods, the library contained DNA from a species-complex, resulting in very few shared loci across 80% of the samples. Therefore, in this case results based on loci shared by 50% of samples are shown. Optimal M = n values were decided in all cases with this information (and reported in Table 5). Note, however, that it is impossible to make absolute calls regarding the ideal value.