Annotation of the chemoreceptor genes
We performed a comprehensive curation of all members of the major chemoreceptor gene families encoded in the genome of D. silvatica , namely the Gustatory-receptor (Gr ) and Ionotropic (glutamate) receptor (Ir/iGluR ) families (Vizueta, Escuer, Frıas-Lopez, et al., 2020; Vizueta et al., 2018). For this task, we used the pipeline BITACORA (Vizueta, Escuer, Sánchez-Gracia, & Rozas, 2020; Vizueta, Sánchez-Gracia, & Rozas, 2020), along with the homologous sequence data set and hidden Markov model (HMM) profiles used in Vizueta et al. (2018) and using the annotated gene models and genome sequence as input. The resulting identified proteins were validated, and re-annotated when necessary, in the Apollo genome browser (Lee et al., 2013). We classified a gene as “complete” if the length of the encoded protein contains, at least, 80% of the protein domain length characteristic of the family (235 and 180 amino acids, for the GR and IR/iGluR proteins, respectively). The remaining incomplete gene models that could not be recovered using Apollo were classified as “partial” fragments. For each chemoreceptor family, we computed the minimum number of chemoreceptor sequences that could be unequivocally attributed to different genes (S MIN) as in Vizueta et al. (2018).