First time user of GSEA greeted by Error Message

[Bogale Aredo]

Hello,

I am a first-time user of the online GSEA for bulk RNAseq data analysis. I was greeted by several error messages and passed most of it by trial and error. Now I am stuck with the following error message (see attached). I also included other important information.

Best regards,

Bogale

[Anthony Castanza]

Hi Bogale,

Sorry for the delayed reply here, your message got caught up in the google groups spam filter for some reason, and I was on vacation and unable to release it until now.

The screenshot you sent of your dataset file looks fine, although it looks like these are raw counts and not normalized counts. You might try running this data through DESeq2 and dumping out the normalized counts matrix instead - this normalization is what we recommend for using RNA-seq data with GSEA. If you're not super familiar with DESeq2 and not comfortable using it at the command line, the GUI wrapped versions on both cloud.genepattern.org, and usegalaxy.org are capable of outputting the normalized counts matrix. The GenePattern implementation will even accept your GCT file and output the normalized counts matrix as a GCT file as well.

As to the specific error message you're encountering however, are you seeing this when loading your GCT file (which should be saved in UTF-8 tab delimited text - which does appear to be correct from the screenshot), with the file extension changed to .gct not .txt, or when loading the CLS file which describes the phenotype-to-sample mappings in the CLS format defined here: https://docs.gsea-msigdb.org/#GSEA/Data_Formats/#phenotype-data-formats

I didn't see a screenshot of a CLS file in your document, so can't assess if there is a formatting issue there. But it should also be UTF-8 plain text, formatted to that linked specification, with the file extension changed to .cls and not .txt.

The selection of chip and gene set database appear to be appropriately matched, so there shouldn't be any issue there.

If you like, you can send your gct and cls files confidentially to gsea...@broadinstitute.org and I can take a look and see if I can figure out what the issue is.

-Anthony

Anthony S. Castanza, PhD

Curator, Molecular Signatures Database

Mesirov Lab, Department of Medicine

University of California, San Diego