Gene co-expression networks identify novel candidate genes for moulting and development in the Atlantic salmon louse (Lepeophtheirus salmonis)

Background The salmon louse ( Lepeophtheirus salmonis ) is an obligate ectoparasitic copepod, living on Atlantic salmon and other salmonids in the marine environment. Salmon lice cause a number of environmental problems and lead to large economical losses in aquaculture every year. In order to develop novel parasite control strategies, a better understanding of the mechanisms of moulting and development of the salmon louse at the transcriptional level is required. Methods Three weighted gene co-expression networks were constructed based on the pairwise correlations of salmon louse gene expression profiles at different life stages. Network-based approaches and gene annotation information were applied to identify genes that might be important for the moulting and development of the salmon louse. RNA interference was performed for validation. Regulatory impact factors were calculated for all the transcription factor genes by examining the changes in co-expression patterns between transcription factor genes and deferentially expressed genes in middle stages and moulting stages. Results Eight gene modules were predicted as important, and 10 genes from six of the eight modules have been found to show observable phenotypes in RNA interference experiments. We knocked down five hub genes from three modules and observed phenotypic consequences in all experiments. In the infection trial, no copepodids with the RAB1A-like gene knocked down were found on fish, while control samples developed to chalimus-1 larvae. Also, a FOXO-like gene obtained highest scores in the regulatory impact factor calculation. Conclusions We propose a gene co-expression network-based approach to identify genes playing an important role in the moulting and development of salmon louse. The RNA interference experiments confirmed the effectiveness of our approach and demonstrated the indispensable role of RAB1A-like gene in the development of salmon louse. In addition to salmon louse, this approach could be generalized to identify important genes associated with a phenotype of interest in other organisms. ### Competing Interest Statement The authors have declared no competing interest. * ### Abbreviations BC : betweenness centrality BLAST : Basic Local Alignment Search Tool BP : biological process CT : threshold cycle CC : cellular component DE : differentially expressed/differential gene expression dsRNA : double-stranded RNA DW : differential wiring fw : forward FWER : family-wise Type I error rate GCN : gene co-expression network GO : gene ontology KD : knock down kIM : intramodular connectivity kME : module eigengene-based connectivity MF : molecular function RIF : regulatory impact factor RNAi : RNA interference RNA-seq : RNA sequencing rv : reverse TF : transcription factor