Can CAMERA analyze just one particular pathway
1
0
Entering edit mode
@200b7413
Last seen 14 months ago
Portugal

Hi!

I have a gene set of 1500 genes from RNA seq from which I did DGE using limma+voom then over the DGE genes I already did ORA using enrichGO analyzing MF, CC and BP:

go_enrich_1 <- enrichGO(gene = genes,
              universe = names(gene_list_1),
              OrgDb = org.Hs.eg.db,
              keyType="SYMBOL",
              ont = "ALL",
              pAdjustMethod = "fdr",
              pvalueCutoff = 0.05,
              readable = TRUE)

And ORA using the Cancer Pathways like this:

C3_t2g <- msigdbr(species = "Homo sapiens", category = "C3") %>% 
dplyr::select(gs_name, gene_symbol)
em <- enricher(genes, TERM2GENE=C3_t2g)

Let's say I would like to analyze just one pathway, more precisely the Glycolisys pathway. Can I use CAMERA method to analyze just one pathway and if yes how would I do it? And if not how can I do this?

Best, Manuel

pathway ORA limma CAMERA DGE • 960 views
ADD COMMENT
1
Entering edit mode
@gordon-smyth
Last seen 5 hours ago
WEHI, Melbourne, Australia

Dear Manuel,

I refered you 5 weeks ago to a workflow using camera() for GSEA analysis (How to do GSEA over limma + voom DGE ?). Did it not work for you? The camera() workflow is the same whether you have one pathway or many.

I'm a little surprised you have used ORA for the MSigDb cancer pathways, because GSEA methods like camera() are generally considered better for this purpose.

Personally, I do tend to use roast() rather than camera() if I am testing for just one pathway in isolation. However one could use camera().

Best wishes
Gordon

ADD COMMENT
0
Entering edit mode

Dear Gordon, hope you are well and thank you for answer.

I decided to do ORA instead of GSEA due to the fact that I have a set of 1500 genes of RNA Seq and not the entire gene set (whole exome). GSEA is not recommended when the RNA dataset does not include the entire gene set.

To use camera I would have to pass to the method a list of genes included in the pathway, in my case, glycolisis and pass a list of the genes filtered by logfc values?

Best Regards, Manuel

ADD REPLY
0
Entering edit mode

I decided to do ORA instead of GSEA due to the fact that I have a set of 1500 genes of RNA Seq and not the entire gene set (whole exome). GSEA is not recommended when the RNA dataset does not include the entire gene set.

I think you may have been incorrectly advised. In reality, camera() and ORA have exactly the same requirements in terms of the background universe of genes.

In any case, if you have RNA-seq, then you have data on all genes.

To use camera I would have to pass to the method a list of genes included in the pathway, in my case, glycolisis

Yes, to use any gene set test you have to identify which genes are in the set.

and pass a list of the genes filtered by logfc values?

No, nothing like that is required. Why would you think it is?

Please read the camera() documentation and examples. I'd be happy to give you help with code, but you have to explain more about your data and at least make an attempt to use camera(). What is your actual data? How is it that you have a set of only 1500 genes? How did you select those 1500 genes? Are you analysing RNA-seq or whole exome? It can't be both!

ADD REPLY

Login before adding your answer.

Traffic: 824 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6