The editor has been updated to markdown! Please see more info at: Tutorial: Updated Support Site Editor

Question: About correlation tests between pairs of genes
gravatar for Jon Bråte
4.3 years ago by
Jon Bråte150
Jon Bråte150 wrote:


I have pairs of transcripts that overlap on the genome, but I am not sure if they actually belong to the same gene or not. I have expression data for many different samples and I was thinking to perform a correlation test between each overlapping pair.

I can do cor.test() with method="spearman", but I guess maybe this is not the best for RNA-seq data? And also I need to correct for multiple testing.

Any suggestions on how to best perform this test? I want to remove all the gene pairs that are significantly correlated.



multtest • 593 views
ADD COMMENTlink modified 4.3 years ago by Simon Anders3.5k • written 4.3 years ago by Jon Bråte150
Answer: About correlation tests between pairs of genes
gravatar for Simon Anders
4.3 years ago by
Simon Anders3.5k
Zentrum für Molekularbiologie, Universität Heidelberg
Simon Anders3.5k wrote:

Hi Jon

I am very puzzled by your question. If you know that the transcripts originate from the same locus, in which sense could they not be from the same gene?

And why would correlation between these transcripts confirm that they belong to the same gene? How have you got your expression measure for these transcripts? If they are from the same region, then they have been quantified using (at least partially) the same reads, and so are bound to appear to correlate.

There might be some circular reasoning here.



ADD COMMENTlink written 4.3 years ago by Simon Anders3.5k

You are correct to be puzzled :)

I should have mentioned that I do not know from which strand the reads transcripts originate from. And so therefore the overlapping transcripts could be on separate strands.

About the expression measurements, they were obtained by mapping RNA-seq reads to the genome, and then counting based on gff-files. And you are right that some of the same reads will have been used to quantify if the transcripts overlap. I need to think about that... Good point!

ADD REPLYlink written 4.3 years ago by Jon Bråte150
Please log in to add an answer.


Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 16.09
Traffic: 210 users visited in the last hour