Create a regulatory network between exosomal and endogenous miRNAs

Question

[diego1530]

Dear all.

By sequencing two libraries, I have data on differentially expressed miRNAs and piRNAs, both in exosome vesicles and of endogenous origin in glial cells.Next, my interest with the expressed miRNAs and piRNAs is to create a regulatory network between those of exosomal and endogenous origin to identify possible co-expression with these sRNAs and already predicted targets. For this purpose, and since I have searched the literature, but I have not found anything about it, I would like to ask if you know or can suggest me any R integrated package that could do this analysis.

Really, I would be deeply grateful for your help.

Answer 1

[Elucidata]

Given you already have the differential expression results, we assume you have properly taken care of steps related to aligning and annotating the small RNAs of your interest. You can use the expression values of your miRNAs and piRNAs across your samples and compute correlations or similar metrics. This should give you an n X n square matrix where n is the number of your miRNAs and piRNAs. This matrix can be used to create a network/graph. A popular R package for doing this is igraph (available from CRAN).

Another approach for doing such analyses is using WGCNA (with helpful tutorials here). The approach and the metrics you choose should be guided by the kind of analysis you wish to do (in your case, identifying co-expression). This 2019 paper benchmarks different metrics which might be of help to you (the original article is pay-walled, the authors have shared a free, public copy here).