GSEA error 1005 "The collapsed dataset was empty when used with chip:ftp...."

Question

[mahejabeen.nidhi]

I am using the GUI version of GSEA. The samples are of mice. I prepared the required files, expression dataset (.gct) and phenotypelabel (.cls), as required.

Expression dataset (partial, feature used are normalized counts):

#1.2                    
48526   4               
NAME    Description CTL_1   CTL_2   CTL_3   Mg_3
Gnai3   NA  5814.957078 6282.049601 7795.119817 4370.093527
Pbsn    NA  0   0   0   0
Cdc45   NA  117.9117583 85.22736483 61.28238849 207.5287133
H19 NA  0   0   0   0
Scml2   NA  0.866998222 0   0   1.844699674

Phenotype label (complete):

4 2 1
# CTL Mg
CTL CTL CTL Mg

Gene sets database:

ftp.broadinstitute.org://pub/gsea/gene_sets/h.all.v7.1.symbols.gmt

Collapse/Remap to gene symbols:

Collapse

Permutation type:

gene_set

Chip platform:

ftp.broadinstitute.org://pub/gsea/annotations_versioned/Mouse_Illumina_MouseRef_8_to_Human_Orthologs_MSigDB.v7.1.chip

The error I am getting is as follows:

<Error Details>

---- Full Error Message ----
The collapsed dataset was empty when used with chip:ftp.broadinstitute.org://pub ...

---- Stack Trace ----
# of exceptions: 1
------The collapsed dataset was empty when used with chip:ftp.broadinstitute.org://pub/gsea/annotations_versioned/Mouse_Illumina_MouseRef_8_to_Human_Orthologs_MSigDB.v7.1.chip------
xtools.api.param.BadParamException: The collapsed dataset was empty when used with chip:ftp.broadinstitute.org://pub/gsea/annotations_versioned/Mouse_Illumina_MouseRef_8_to_Human_Orthologs_MSigDB.v7.1.chip
    at org.gsea_msigdb.gsea/xtools.gsea.Gsea.getDataset(Gsea.java:95)
    at org.gsea_msigdb.gsea/xtools.gsea.Gsea.execute(Gsea.java:141)
    at org.gsea_msigdb.gsea/edu.mit.broad.xbench.tui.TaskManager$ToolRunnable.run(TaskManager.java:435)
    at java.base/java.lang.Thread.run(Unknown Source)

Could someone please help me understand what could be causing this error? I was able to run the analysis on human samples using a similar approach.

Answer 1

[Anthony Castanza]

You've selected the wrong chip file for your dataset, the chip you selected is for a specific Illumina Mouse Microarray. Your data is Mouse Gene Symbols, so the correct chip is Mouse_Gene_Symbol_Remapping_to_Human_Orthologs_MSigDB.v7.1.chip

I would also suggest that you use our help form at https://groups.google.com/group/gsea-help for getting help with GSEA errors, you can ask us your questions directly, and we endevor to answer as quickly as possible.

I should also note that it looks like you only have one sample for one of your phenotypes. Most of GSEA's methods require replicates. You're not going to be able to use the default Signal2Noise ranking metric (or TTest). You might consider ranking your genes outside of GSEA and providing that rnk list to GSEA Preranked.

-Anthony

Anthony S. Castanza, PhD

Curator, Molecular Signatures Database

GSEA-MSigDB Team