Question: how to get rid of redundancies in an RNA-seq experiment but preserving genes changing in opposite directions?
0
gravatar for Mozart
3 months ago by
Mozart120
Mozart120 wrote:

Hello there, it ha been a while since my last post. I spent a lot of time with my music stuff (please, see my nickname if you don't have sarcasm), I know you missed me!

today, I would like to ask you a very simple question that I was not able to find in any of the other posts over there (mea culpa).

I have an RNA-seq experiment in which I have to find all of the genes that changed upon a specific treatment. We know that this effect should be reconnected to a specific receptor that is expressed on the cell surface. As we designed the experiment, we treated the WT and KO cells with the ligand that specifically recognise the above mentioned receptor. We did this because we wanted to be sure that all of the genes changed in the WT treated vs WT untreated were directly linked to this receptor; so that, if I wanted to check that a certain gene is context specific, I would compare the KO condition to the treated (or untreated) condition and take into account all of the genes changed significantly as a proof that those genes are potentially interesting for my analysis. so, in this case I have three conditions

1(ko vs wt-treat) 2(ko vs wt-untreat) and 3(wt-treat vs wt untreated)

let's say that in (wt-treat vs wt untreated) I find that gene A is up-regulated [i.e. the treatment increase the transcription of the latter one], how can I know that this this change is receptor-dependent? I would expect an opposite change (or a down regulation in this case) of gene A in ko vs wt-treat and virtually an absence of this gene in ko vs wt untreated.

First of all, I just wanted to be sure that everyone (especially the big names here!) agrees with me here. Is this rationale correct for you, guys? if this is correct I may need a help to find a way in R for removing all of the genes that change in the same way in case we are considering (ko vs wt-treat) and are found to be differentially expressed (either ways I guess?) in (wt-treat vs wt untreated).

I am sorry I am genuinely asking you a code here, but I was away from Studio for too long and spent too much time in the lab! But I guess the answers here may be of help for everyone in my similar case.

with love, Mozart

rna-seq data mining • 203 views
ADD COMMENTlink modified 3 months ago by JC7.8k • written 3 months ago by Mozart120
1
gravatar for JC
3 months ago by
JC7.8k
Mexico
JC7.8k wrote:

I can say everything depends on what are you looking for, in my opinion:

  1. wu-treat vs wu-untreat will reveal how your cells respond to the treatment normally
  2. ko-untreat vs wu-untreat will reveal if your KO mutant affects the normal condition, i.e. if the gene is expressed in normal conditions, the KO will reveal the gene as missing and some other affected genes
  3. ko-treat vs wu-treat will reveal how the KO-gene affects the response to the treatment
  4. ko-treat vs ko-untreat will reveal how your mutant cells respond to the treatment

All comparison can be done with edgeR from Bioconductor.

ADD COMMENTlink modified 3 months ago • written 3 months ago by JC7.8k
Please log in to add an answer.

Help
Access

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.3.0
Traffic: 1967 users visited in the last hour