Project description:Transcriptome resources for social insects have the potential to provide new insight into polyphenism, i.e., how divergent phenotypes arise from the same genome. Here we present a transcriptome based on paired-end RNA sequencing data for the ant Formica exsecta (Formicidae, Hymenoptera). The RNA sequencing libraries were constructed from samples of several life stages of both sexes and female castes of queens and workers, in order to maximize representation of expressed genes. We first compare the performance of common assembly and scaffolding software (Trinity, Velvet-Oases, and SOAPdenovo-trans), in producing de novo assemblies. Second, we annotate the resulting expressed contigs to the currently published genomes of ants, and other insects, including the honeybee, to filter genes that have annotation evidence of being true genes. Our pipeline resulted in a final assembly of altogether 39,262 mRNA transcripts, with an average coverage of >300X, belonging to 17,496 unique genes with annotation in the related ant species. From these genes, 536 genes were unique to one caste or sex only, highlighting the importance of comprehensive sampling. Our final assembly also showed expression of several splice variants in 6,975 genes, and we show that accounting for splice variants affects the outcome of downstream analyses such as gene ontologies. Our transcriptome provides an outstanding resource for future genetic studies on F. exsecta and other ant species, and the presented transcriptome assembly can be adapted to any non-model species that has genomic resources available from a related taxon.
Project description:Phenotypic variation arises from interactions between genotype and environment, although how variation is produced and then maintained remains unclear. The discovery of the nest-mate recognition system in Formica exsecta ants has allowed phenotypic variation in chemical profiles to be quantified across a natural population of 83 colonies. We investigated if this variation was correlated or not with intrinsic (genetic relatedness), extrinsic (location, light, temperature), or social (queen number) factors. (Z)-9-Alkenes and n-alkanes showed different patterns of variance: island (location) explained only 0.2 % of the variation in (Z)-9-alkenes, but 21-29 % in n-alkanes, whereas colony of origin explained 96 % and 45-49 % of the variation in (Z)-9-alkenes and n-alkanes, respectively. By contrast, within-colony variance of (Z)-9-alkenes was 4 %, and 23-34 % in n-alkanes, supporting the function of the former as recognition cues. (Z)-9-Alkene and n-alkane profiles were correlated with the genetic distance between colonies. Only n-alkane profiles diverged with increasing spatial distance. Sampling year explained a small (5 %), but significant, amount of the variation in the (Z)-9-alkenes, but there was no consistent directional trend. Polygynous colonies and populous monogynous colonies were dominated by a rich C23:1 profile. We found no associations between worker size, mound exposure, or humidity, although effect sizes for the latter two factors were considerable. The results support the conjecture that genetic factors are the most likely source of between-colony variation in cuticular hydrocarbons.