How to get negative enrichment score

[Anna Huang]

I am looking for advice on the issue of not getting negative enrichment scores from my GSEA analysis. I am using desktop GSEA4.4.0 with GSEAPreranked. I had do Collapse gene symbols due to duplicate matching. 

Below is my analysis:

Gene sets database, C5.hpo

Number of permutations, 1000

Rank list, I ranked my data with directional p value

Collapse, I collapsed due to duplicated matching 

Chip platform: mouse ensemble gene id

Any tip is much appreciated!

[Anthony Castanza]

Hi Annabelle,

Our apologies, your messages were initially caught in the google groups spam filter seemingly for no particular reason.
To clarify; your ranking metric is the sign of the logFC*-log10(pValue), yes?
And with this ranking metric, your data only has positive enrichment results, and you looked in the separate report sections for each phenotype (pos and neg)?

Would you mind posting the enrichment plot from one of the sets that you did obtain a result from? I might help in assessing any dataset distributional issues that could be giving rise to this issue.

It can just simply happen sometimes though if there aren't sets in your analysis that are a good fit for the genes that are down-regulated, but it's pretty rare to see nothing at all. Any additional information you can provide us about the experimental design or GSEA configuration would be helpful.

-Anthony

Anthony S. Castanza, PhD

Curator, Molecular Signatures Database

Mesirov Lab, Department of Medicine

University of California, San Dieg

[Anna Huang]

Hi Anthony, 

Thanks for the quick response. 

1. Yes, my ranking metric is the sign of the logFC*-log10(pValue)

2,  You are right. It turned out that I missed the neg section and they are actually there!

Thank you so much again. 

Anna

--
You received this message because you are subscribed to the Google Groups "gsea-help" group.
To unsubscribe from this group and stop receiving emails from it, send an email to gsea-help+...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/gsea-help/797ce264-8414-4106-9a05-a869d499a0dfn%40googlegroups.com.

[Bogale Aredo]

Hello Anthony,

I am a new user of GSEA. I also use desktop GSEA 4.4. I run into several error messages at several steps. After few trials and error, I moved forward but stuck this one:  col:1 > matrix's fColCnt:1 (see attached).

I am using mouse tissue Bulk RNA seq normalized data per sample per group.  

I selected m5.all.v2025 for gene sets database

I selected Mouse_gene_symbol_Remapping for Chip platform

Are these correct matches?

Best regards,

Bogale