Whole Brain Transcriptome for S. ocellatus

De novo transcriptome assembly was performed using Trinity (Haas et al. 2013) using RNA-Seq reads from male (sneaker, satellite, and nesting male) and female S. ocellatus whole brain mRNA. Lowly-abundant transcripts with an FPKM < 0.5 were filtered by RSEM, resulting in an assembly with 80,103 contigs with an N50 of 2119. To annotate the S. ocellatus transcriptome we employed Blastx (using an E-value cutoff of 1X10-10) on protein sequences for Oryzias latipes and Takifugu rubripes obtained from the Ensembl BioMart database, which resulted in the alignment of 25,856 contigs to the O. latipes genome and 25,213 contigs to the T. rubripes genome. Contigs that aligned with high confidence to more than one database were annotated based on the lowest E value and highest % identity to either reference genome, resulting in an assembly where roughly 32% of high quality contigs were successfully annotated. Contigs without annotations are labelled with contig tags generated by Trinity while annotated contigs have gene symbol following contig tags.