2.11 Gene ontology (GO) enrichment analysis
A R package “org.Dalabrax.eg.db” specific to sea bass annotation was generated using the MakeOrgPackage function from the AnnotationForge Package v1.40.1. The R package ClusterProfiler version 4.6.0 (Yu et al. 2012) with the “org.Dalabrax.eg.db” was used to analyze functional profiles of differentially methylated and expressed genes and to identify major biological functions. The functional analysis was performed using as input the differentially methylated genes (DMGs), differentially methylated promoters (DMPs) and differentially expressed genes (DEGs). A hypergeometric test was performed and enrichment p-value of gene ontology was calculated to find significantly enriched GO terms in the input list of DMGs, DMPs and DEGs. A p- value of <0.05 was set as the threshold value. DMGs, DMPs and DEGs were categorized in three categories belonging to the main GO ontologies: biological process (BP), molecular function (MF) and cellular component (CC). In order to study the differences in functional annotation according to DNA methylation, we separated the DMGs and the DMPs into 4 groups: hyper-methylated gene bodies in FW- vs SW-acclimated fish (GB hyper), hypo-methylated gene bodies (GB hypo), hyper-methylated promoters (PR hyper) and hypo-methylated promoters (PR hypo). We proceeded in the same way for the transcriptome by indicating genes overexpressed in FW (up FW) or in SW (up SW). The 5 GO terms of each analysis with the most significant p-values have been represented.
In order to highlight a functional link between changes in methylation and expression patterns between salinity conditions, we identified the GO terms from the methylome and transcriptome enrichment analyses. Enriched GO terms between DEGs and DMRs (in genes and promoters) were extracted and used to generate three Venn diagrams, one per GO term category: BP, MF, and CC. This enabled us to identify enriched GO terms that are common to both analyses (methylome and transcriptome).