Les multiplexed per lane, randomly distributed across four lanes.Mikheyev and Linksvayer.eLife ;e..eLife.ofResearch articleGenomics and evolutionary biologySequences had been postprocessed by cutadapt (Martin,) to take away Illumina adapter sequences and ConDeTri (Smeds and Kunstner,) to eliminate lowquality bases.Reference genome sequencing and assemblyDNA from a single haploid male ( ng) was utilised to prepare a TruSeq library, which was sequenced in multiplex on an Illumina HiSeq , yielding ,, million bp read pairs.Raw genomic reads have been excellent and adaptor trimmed utilizing ConDeTri and cutadapt (Martin, Smeds and K unstner,), generating ,, read pairs and ,, single reads (.Gb total).The assembly was carried out applying ABYSS, using a range of kmers from to (Simpson et al).We then chose the assembly using the longest N because the reference for transcriptome assembly.Genome assembly quality was evaluated applying the CEGMA pipeline (Parra et al), and by remapping the paired end trimmed reads making use of bowtie (Langmead and Salzberg,).Referencebased transcriptome assembly, annotation and differential gene 5′-?Uridylic acid Endogenous Metabolite expression analysisThe transcriptome was mapped towards the reference applying Tophat , and assembled into transcripts using Cufflinks .(Roberts et al Kim et al).Gene expression information were obtained by remapping the transcript reads to the extracted transcripts utilizing RSEM and calculating the expected counts in the gene level (Li and Dewey,).When various isoforms of a single locus have been located, only the longest transcript was made use of for gene annotation.Assembled transcripts had been annotated employing BLASTX in the nonredundant NCBI database with expectation values of E .These results had been employed to assign Gene Ontology (GO) profiles with Blastgo (Conesa et al).Differential gene expression evaluation and transcriptional network analysisTranscript counts have been filtered by abundance, removing these with significantly less than fragment per kilobase mapped (FPKM) in much more than half on the libraries (Mortazavi et al).Differential gene expression analysis was carried out in edgeR, using a GLM match for the count data and identifying differentially expressed genes working with planned linear contrasts (Robinson et al).So as to infer coexpression modules and gain an insight into network structure of gene interactions, we performed a weighted gene coexpression network evaluation (WGCNA) around the count information (Langfelder and Horvath,).WGCNA was performed on the complete transcript set, after filtering out the lowabundance transcripts.This analysis relies on patterns of gene coexpression, but has been shown to reconstruct PubMed ID:http://www.ncbi.nlm.nih.gov/pubmed/21487883 protein rotein interaction networks with affordable accuracy (Zhao et al Allen et al).We utilized total connectivity as a measure of gene interaction strength, since it isn’t as sensitive to module assignments, and probably reflects the overall selective pressures acting around the gene, beyond these imposed by its part in age polyethism.As with most gene expression analysis, WGCNA supplies much better estimates for hugely abundant genes, and in particular for genes showing variation in their expression levels.Consequently, lowabundance and invariant genes will show lower connectivity.GO term enrichment analysis was performed working with the R package GOstats (Falcon and Gentleman,).We report GO terms as enriched when p .Evolutionary rate and gene expression conservation analysesFire ant (S.invicta) orthologs for each gene were determined making use of reciprocal finest BLASTP, working with cutoffs of .This parameterization allowed for.