SNP annotation and gene ontology
To obtain functional annotations for candidate loci, we used the LastZ pairwise alignment tool v1.04.00 (R. S. Harris, 2007)⁠ to find homologous M. ochrogaster positions. To do this, we used a 20,000 bp scaffold surrounding each candidate locus as a query in LastZ and the default options for calculating pairwise alignments. Only alignments with a bit score greater than 1,000 and a query coverage of at least 50% were considered. If multiple alignments passed this filter, the alignment with the longest length and highest bit score was selected as the best match. If the loci were on protein-coding regions, we used the UniProt database to examine gene function and find gene ontology (GO) terms. We performed an enrichment analysis using topGO (Alexa & Rahnenführer, 2010) in the ”biological processes” category. We compared our list of candidate genes with all genes with a SNP. We used Fisher’s exact test and the elim algorithm to account for correlation in the topology of the GO graph, and reported the GO terms with a p-value < 0.01 and at least 4 associated genes (Table S6).