Question: Assesing Gene topology and connectivity
gravatar for v82masae
4 months ago by
v82masae130 wrote:

Hi everyone in the Biostar community,

I have been working on detecting Differentially Expressed (DE) genes from a RNAseq experiment between groups, and also assesing Differentially Disperse (DD) genes. Say, contrasting differences between means and between variances. I habe also calculated de Biological Coefficient of Variation (BCV) for each gene and now I want to extract some connectivity and pathway info from these data.

Basically, I have three lists en each group comparison (three group vs group comparisons so 9 lists): one list with genes that are DE and DD at the same time. Another list with genes only DE. And lastly, one list with genes only DD.

I have removed lowly expressed genes (below 3 overall mean CPM) considered as noise, that tend to be very variable (high BCV), and only kept genes DE and or DD with |FC| > 1.5 and FDR < 0.05.

I have aproximately 2000 genes DD and DE, a d 200-300 genes only DD or only DE, in each of the three comparisons, to show some numbers. I have used the recently published MDseq pipeline for getting such results.

So, further ahead, now we would like to try extracting some meaningful biological insights for such DD phenomena. Our hipothesis is that there are some genes that in control conditions tend to be either stable or very variable across samples within one group, and that they experience a change in their dispersion, stabilizing or destabilizing their expression levels in response to our treatment stimulus.

We want to assess what is the centrality in pathways, their connectivity with other genes and coexpresion phenomena that may occur. Our hypothesis here is that some DD genes can trigger other genes to be also DD or even DD and DE at the same time, acting as expression dysregulators.

I havent worked before in such enrichment, coexpression and pathway analyses and I would like to know if you could suggest some pipelines or softwares I could check (mainly R or web-based and open-source).

I have the expression matrices for all genes and also the lists with DD/DE, DE and DD genes.

For coexpresion analyses I have checked WGCNA or PCIT, but I am not sure if these softwares are what I need.

For pathway and connectivity I have hear about igraph package for representing nets but I have not gone any further than that.

Could you give me some advice?

Many thanks

ADD COMMENTlink modified 4 months ago • written 4 months ago by v82masae130
Please log in to add an answer.


Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.3.0
Traffic: 1674 users visited in the last hour