I am trying to run EGSEA un Galaxy. I already have my list of differentially expressed genes, run by Limma-Voom. I have 6 samples in total, 3 controls and 3 treated ones. According to the step-by-step tutorial on your website Hands-on: 3: RNA-seq genes to pathways / 3: RNA-seq genes to pathways / Transcriptomics I created my EGSEA counts and anno files. I did not generate any factordata files though since I’m not dealing with additional factors (batch effects and etc). I set the settings as instructed in the tutorial and certified that I wasn’t going to use the tool for commercial purposes. However, I cannot run the tool at all. There are even no error appearing for me to understand what the problem is. I click on run tool, the work in progress circle appears for half a second and the run tool buttons turns blue again, as if I never hit the run.
To set up the tool with enough information, you’ll need to provide the “group to sample” associations. With a count matrix input choice, this data would come from a factor information file. Otherwise, the groups referenced in the contrasts of interest cannot be mapped back to samples in the count matrix! (I hope that wasn’t too confusing!).
The solution is to do one of these:
input the count files separately instead of combined in a matrix. The tool form will have areas where you can directly type in the top level single factor label and the group labels
or create a factor file with the samples, group, and factor information
If you need help creating a factor file, please let us know. Seeing a peek of the header in your count matrix dataset will help to get this started!
More about what is going wrong. There is a technical issue with the usage alerts on the form. Tool forms will block the submission function when the minimum required information is not input on the form (this was working) but the usual method of alerting about the “missing input data” is hidden (could be improved). Specifically, the form should automatically toggle the “Input factor information from file?” option to a required Yes whenever the option for Single Count Matrix is used. This seems familiar to me, and may already be requested, but I’ll follow up and post back an update here.
