Outlier Gene Annotation
We obtained consensus sequences for each outlier locus using thepopulations program of STACKS. Consensus sequences were compared to NCBI’s GenBank nucleotide database (NCBI Resource Coordinators 2018), and the N. lapillus transcriptome (Chu et al. 2014) using the BLASTN 2.12.0+ (Camacho et al. 2009) function to identify closely matching sequences that may provide information on locus identity. Matches to NCBI’s GenBank nucleotide database were filtered to include only those with percent identities between 70-100% and E-values between 0 and 1E-06, and the top matching sequence (based on E-value) for each locus was recorded. From these results, we identified loci that were annotated to coding regions. Matches to the N. lapillustranscriptome were annotated by comparing transcriptome contig sequences that matched an outlier locus to the UniProtKB Swiss-Prot database (The Uniprot Cosortium 2021) using BLASTN, with percent identity and E-value cutoffs of 50% and 1E-06, respectively.