Computational protocol: Integrative testis transcriptome analysis reveals differentially expressed miRNAs and their mRNA targets during early puberty in Atlantic salmon

Similar protocols

Protocol publication

[…] 3′-UTR’s sequences for salmon mRNAs were extracted as the sequence spanning from CDS stop to the end of transcripts using a genbank flat file hosted at ncbi (GCF_000233375.1_ICSASG_v2_rna.gbff.gz). In cases where there were multiple annotations for a transcript, the longest 3′-UTR was retained. Sequences for each 3′-UTR was then compiled to a multi fasta file. Based on this salmon 3′-UTR reference multi fasta file, targets for differentially expressed miRNAs were predicted with miRMap []. Expression correlation between miRNAs and miRNA targets was calculated based on expression profiles from miRNAs as well as mRNAs from the same samples: An expression profile for averaged measurements across stages for each miRNA & mRNA was made. Then using the python pandas.expanding_corr function, a pairwise Pearson correlation coefficient (1: perfectly correlated, −1: anticorrelated) was calculated for these expression profiles. For downstream pathway analysis only mRNAs supported by a predicted miRNA-mRNA relation (based on the 20% top scoring miRmap target predictions for each miRNA) and with an anti-correlated expression (Pearson < −0.5), were used as input for KEGG analyses using the R ClusterProfiler package []. Target predictions and associated pathways were then visualized in Cytoscape []. In order to prevent excessive cluttering of the network, only a subset of these target relations consisting of the top 50% scoring miRNA predictions for each pathway and with a negative expression correlation less than −0,8 were visualized. Nodes of the network (genes) were colored according to the strength of the negative correlation with their respective targeting miRNAs (Fig. ). […]

Pipeline specifications

Software tools miRmap, clusterProfiler
Databases KEGG
Application Genome annotation