Closed yaaminiv closed 5 years ago
Hi Yaamini - thanks for trying out my method!
This error usually pops up when some formatting of the input file(s) is wrong - open you “measures” file in a plain text editor to confirm that “measures” is indeed a comma-separated table with no additional things, and “annotation” is tab-delimited, with multiple GO terms separated by just semicolon (no spaces!)
One thing though - you seem to have included only “significant" genes in the measures table. This is wrong - you should list all genes for which the testing was done, significant as well as non-significant ones. This is a particular strength of GO_MWU - to does not require any prior statistical testing or imposing of arbitrary “significance cutoffs” instead it looks at the whole list of genes ranked by response strength. You can even use raw log-fold-changes as a measure to rank them.
Misha
On Jul 12, 2019, at 3:14 AM, Yaamini Venkataraman notifications@github.com wrote:
I am running my modified version of a GO-MWU script (found here https://github.com/eimd-2019/project-EWD-transcriptomics/blob/master/analyses/GO-MWU/nonZostera-GO_MWU.R), using this annotation file https://github.com/eimd-2019/project-EWD-transcriptomics/blob/master/analyses/GO-MWU/2019-07-11-nonZostera-GO-Annotations-Table.tab and this table of significance measures https://github.com/eimd-2019/project-EWD-transcriptomics/blob/master/analyses/GO-MWU/2019-07-11-nonZostera-Table-of-Significance-Measures.csv. I keep encountering the following error when running gomwuStats:
https://user-images.githubusercontent.com/22335838/61109451-c4998f80-a439-11e9-8926-9ce320c50e99.png https://user-images.githubusercontent.com/22335838/61109453-c6635300-a439-11e9-9b5d-6a05ff7a4ef0.png The specific input files were created with this script https://github.com/eimd-2019/project-EWD-transcriptomics/blob/master/analyses/GO-MWU/2019-07-11-Gene-Enrichment-with-GO-MWU.Rmd. I ran a similar script https://github.com/eimd-2019/project-EWD-transcriptomics/blob/master/analyses/GO-MWU/Zostera-GO_MWU.R successfully using an annotation table https://github.com/eimd-2019/project-EWD-transcriptomics/blob/master/analyses/GO-MWU/2019-07-11-Zostera-GO-Annotations-Table.tab and significance table https://github.com/eimd-2019/project-EWD-transcriptomics/blob/master/analyses/GO-MWU/2019-07-11-Zostera-Table-of-Significance-Measures.csv generated in the same script, so I'm unsure why I'm getting this error now.
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/z0on/GO_MWU/issues/3?email_source=notifications&email_token=ABZUHGE3ZNC2HNODNSZF6BDP7AVOXA5CNFSM4ICDMY22YY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4G6ZZ2MA, or mute the thread https://github.com/notifications/unsubscribe-auth/ABZUHGFCJ3UTDMDESZCWQWLP7AVOXANCNFSM4ICDMY2Q.
This error usually pops up when some formatting of the input file(s) is wrong - open you “measures” file in a plain text editor to confirm that “measures” is indeed a comma-separated table with no additional things, and “annotation” is tab-delimited, with multiple GO terms separated by just semicolon (no spaces!)
I deleted my original files and used the same code I had to recreate them. Not sure what the difference was this time, but it worked!
you should list all genes for which the testing was done, significant as well as non-significant ones
Re-ran the code using all genes. Thank you for the clarification!
I am running my modified version of a GO-MWU script (found here), using this annotation file and this table of significance measures. I keep encountering the following error when running
gomwuStats
:I ran a similar GO-MWU analysis successfully using an this annotation table and this significance table, so I'm unsure why I'm getting this error now. Both sets of input files were created with this script.
Is there any sort of code modification I need to make in order to run the script?