Question: KEGG enrichment for non-model organism
0
gravatar for sbchua.1990
6 months ago by
sbchua.199010
sbchua.199010 wrote:

Hi, just want to know if I have obtained KEGG annotation and mapped to pathway using KEGG mapper (http://www.genome.jp/kegg/mapper.html). Due to my lack of skill in doing the gene set enrichment analysis, I have think of a simpler way but not sure whether it is correct?

I simply tabulate how many genes in the mapped to particular gene set (pathway) for both background (whole transcriptome assembly) and sample (DEGs). And also tabulate total genes annotated for both background (whole transcriptome assembly) and sample (DEGs).

Then I will perform using hypergeometric test tool available online (https://www.geneprof.org/GeneProf/tools/hypergeometric.jsp) for every individual pathway.

Is this the correct way? Any simpler way to do this? Or if I am wrong, please guide me to the correct path.

rna-seq R gene • 499 views
ADD COMMENTlink written 6 months ago by sbchua.199010
1

What you're doing is not entirely clear to me. If you've got a contingency table, the standard way to test for enrichment is using Fisher's exact test.

ADD REPLYlink written 6 months ago by Jean-Karim Heriche14k

I agree with @Jean-Karim Heriche. Fisher is usually the recommended option. I had great satisfactions using the enrichKEGG function of the R Bioconductor package clusterProfiler.

ADD REPLYlink modified 6 months ago • written 6 months ago by Fabio Marroni1.6k

Thanks for reply, will clusterProfiler work for non-model species? Can I choose to use my own whole assembly as background?

For simplicity, let say I have 10,000 genes assembled. 2,000 were annotated to at least one pathway (for background). I have 1,000 DEGs, 180 annotated to at least one pathway. For 'pathway A', 100 genes were found in background and 10 were found in DEGs.

So my contingency table should be something like A (10, 180-10, 100, 2000-100) or B (10, 1000-10, 100, 10000-100). I am not sure whether to include the genes that are not annotated to any pathway.

ADD REPLYlink written 6 months ago by sbchua.199010

I'm looking for the answer to a similar question. Were you able to use your own assembly as background in clusterProfiler?

ADD REPLYlink written 4 months ago by k_mer0

How did you solve your problem?

ADD REPLYlink written 3 months ago by Ric180
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: 1586 users visited in the last hour