Building EnrichmentMap error

[amkhasawneh]

Hello.

Thanks a lot for all your efforts.

Whenever I try to build an enrichment map, I get a strange error:

How should I go about this?

Thanks a lot. Abdullah

[risserlin]

Often that error is associated with specifying the wrong analysis type in the enrichment input window for the enrichment analysis results file. Are you trying to run a GSEA or a generic analysis?

[amkhasawneh]

Thank you.

I'm trying to run a GSEA analysis, which I did with the fgsea R package. I have divided the output file into two--one with positive and one with negative NESs.

[risserlin]

Are you just giving Enrichment map the file that fGSEA generates? Unfortunately EM is not configured to recognize the fgsea files directly.

[risserlin]

Sorry not pretty and specific to my current analysis but I have a function like this in R to create GSEA like results from fGSEA outputs.

write_patient_fgsea_results<- function(current_fgsea_results, current_results_dir, 
                                       current_patient){

  current_patient <- substring(current_patient,1,12)

    current_patient_directory_fullpath <- file.path(current_results_dir, 
                                                    current_patient)

    if(!dir.exists(current_patient_directory_fullpath)){
      dir.create(current_patient_directory_fullpath)
    }

    #calculate the rank at max
    #fgsea returns the leading edge.  Just need to extract the highest rank from 
    # set to get the rank at max
    calculated_rank_at_max <- apply(current_fgsea_results,1,
                                    FUN=function(x){ 
                                      max(which(names(current_ranks)
                                                %in% unlist(x[8])))})

    enr_current_patient <- cbind(current_fgsea_results$pathway,
                                     current_fgsea_results$pathway,
                                     "Details",
                                     current_fgsea_results$size,
                                     current_fgsea_results$ES,
                                     current_fgsea_results$NES,
                                     current_fgsea_results$pval,
                                     current_fgsea_results$padj,
                                     0,
                                     calculated_rank_at_max,
                                     apply(current_fgsea_results,1,
                                           FUN=function(x){paste(unlist(x[8]),collapse=",")})) 

    colnames(enr_current_patient) <- c("name","description",
                                           "GS details","SIZE","ES",
                                           "NES","pval","padj","FWER",
                                           "Rank at Max","leading edge genes")

    enr_filename_positive <- paste0(current_patient, "fgsea_enr_results_pos.txt",sep="")
    enr_filename_docker_positive <- file.path(current_patient_directory_fullpath,
                                                  enr_filename_positive)

    write.table(enr_current_patient[which(enr_current_patient[,"NES"]>=0),] ,
                enr_filename_docker_positive,
                col.name=TRUE,sep="\t",row.names=FALSE,quote=FALSE,fileEncoding="latin1")

    enr_filename_negative <- paste0(current_patient, "fgsea_enr_results_neg.txt",sep="")
    enr_filename_docker_negative <- file.path(current_patient_directory_fullpath,
                                                  enr_filename_negative)

    write.table(enr_current_patient[which(enr_current_patient[,"NES"]<0),] ,
                enr_filename_docker_negative,
                col.name=TRUE,sep="\t",row.names=FALSE,quote=FALSE,fileEncoding="latin1")

    #create a  expression file - just the ranks file as expression file
    exp <- data.frame(name = names(current_ranks), 
                          description = names(current_ranks),current_ranks)
    exp_filename <- paste0(current_patient,"expression.txt",sep="")
    exp_name_docker <- file.path( current_patient_directory_fullpath,exp_filename)

    write.table(exp,
                exp_name_docker,
                col.name=TRUE,sep="\t",row.names=FALSE,
                quote=FALSE,fileEncoding="")

    #create a rank expression file
    rnk <- data.frame(name = names(current_ranks), 
                          current_ranks)
    rnk_filename <- paste0(current_patient,"ranks.rnk",sep="")
    rnk_name_docker <- file.path( current_patient_directory_fullpath,rnk_filename)

    write.table(rnk,
                rnk_name_docker,
                col.name=TRUE,sep="\t",row.names=FALSE,
                quote=FALSE,
                fileEncoding="")

}

[amkhasawneh]

Thanks a lot for taking the time, and I'm sorry for the late response (it's because of the time difference).

I have been adding ranks to my fGSEA results, but I'm not sure about the other modifications. I will adapt the function to my data, and try it again with EnrichmentMap.

Thanks again.

[amkhasawneh]

It works perfectly.

Thank you very much.