Question: How to check if specific mutations are or not enriched in a RNAseq data seq?
0
gravatar for Lila M
2.6 years ago by
Lila M 800
UK
Lila M 800 wrote:

Hi everybody, I would like to know if there is any way to analyze if a specific mutation (let's say p53 )is or not enriched in a data set. Is the first time that I try to solve this question but I don't know how to address this issue exactly. I'm reading about GSAR (Gene Set Analysis in R) but not sure if it could cover this.

Any idea?

Thanks!

rna-seq enriched mutation • 726 views
ADD COMMENTlink modified 2.6 years ago by Istvan Albert ♦♦ 82k • written 2.6 years ago by Lila M 800
1

Shouldn't you do variant calling to identify mutations? I'm not following the logic in this thread.

ADD REPLYlink written 2.6 years ago by WouterDeCoster43k

GSAR is for gene set networking, PPI, and other analyses relating to differentially expressed, or candidate genes. Can you clarify if you're looking for something at the gene level, or your looking at point mutations within genes?

ADD REPLYlink written 2.6 years ago by andrew.j.skelton735.9k

Hi, I am looking for something at the gene level. In more detail I have two groups and I would like to know if in for example the mutation p53 is enriched or not in any of the groups.

ADD REPLYlink written 2.6 years ago by Lila M 800
2
gravatar for Istvan Albert
2.6 years ago by
Istvan Albert ♦♦ 82k
University Park, USA
Istvan Albert ♦♦ 82k wrote:

If you know which single gene you are testing for then you are in luck as you can apply a much simpler statistical test (no multiple-test correction is needed).

You could apply a "regular" differential expression test but then use the p-value column rather than the adjusted p-value for this gene. Or you could directly apply a binomial test on the observed counts.

ADD COMMENTlink written 2.6 years ago by Istvan Albert ♦♦ 82k

Thank you very much Istvan, What I have is a gene set derived from differential expression analysis (HTSeq) and DEU for two different treatment. I would like to know if p53 mutations is enriched or not in this two different groups. Can you please give me an example (or some literature) about how to do this kind of analysis? Thank you very much

ADD REPLYlink written 2.6 years ago by Lila M 800
1

HTSeq produces the counts and I don't know what format is your differential expression is shown as but typically it has columns with average expressions for each condition, a p-value, and an adjusted p-value. In this case, since you are only looking at one gene that you've selected apriori, you can use the p-value column. That p-value indicates the likelihood of observing the change of the reported magnitude by chance alone.

There is no literature that you would need to consult to be allowed to make this choice. The question is whether did the method select the "interesting" gene out of all possible options or did you select the gene beforehand. If a method produces the "interesting" gene by looking at all possible options then you have to correct (adjust the pvalue) for multiple comparisons. Otherwise, you don't.

ADD REPLYlink modified 2.6 years ago • written 2.6 years ago by Istvan Albert ♦♦ 82k

Thank you for the explanation! I will do it!

ADD REPLYlink written 2.6 years ago by Lila M 800
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: 886 users visited in the last hour