moved utility cp functions into commons

common/gsea_enrichment.R

 #!/usr/bin/env Rscript
 #+ echo=FALSE, error=F
 
-
-# cd
-
 suppressMessages(require(docopt))
 
 ## todo use textual input here for ease of use
 doc <- '
 Perform an enrichment analysis for a set of genes
-Usage:gsea_enrichment.R[options] <sorted_gene_lists_tsv> <group_col>
+Usage:gsea_enrichment.R[options] <sorted_gene_lists_tsv>
 
 Options:
 --list_id_col Column containing the grouping variable to speparate different lists [default: ]
@@ -23,10 +20,11 @@ opts <- docopt(doc, commandArgs(TRUE)) ## does not work when spining
 
 devtools::source_url("https://raw.githubusercontent.com/holgerbrandl/datautils/v1.27/R/core_commons.R")
 devtools::source_url("https://raw.githubusercontent.com/holgerbrandl/datautils/v1.27/R/ggplot_commons.R")
-devtools::source_url("https://raw.githubusercontent.com/holgerbrandl/datautils/v1.27/R/bio/diffex_commons.R")
+## todo relase and use proper version here
+devtools::source_url("https://raw.githubusercontent.com/holgerbrandl/datautils/master/R/bio/diffex_commons.R")
 
 loadpack(knitr)
-loadpack(DT)
+#loadpack(DT)
 
 #loadpack(clusterProfiler)
 
@@ -41,7 +39,8 @@ loadpack(DT)
 
 ## load the data
 geneLists <- read_tsv(opts$sorted_gene_lists_tsv)
-group_col=opts$group_col
+#group_col=opts$group_col
+group_col="contrast"
 geneLists %<>% group_by_(.dots=group_col)
 
 #geneLists %<>% filter(cluster %in% c("cluster_1", "cluster_2"))
@@ -68,46 +67,10 @@ vec2df(unlist(opts)) %>% filter(!str_detect(name, "^[<-]")) %>% kable()
 listLabels <- geneLists %>% select(-ensembl_gene_id) %>% distinct
 listLabels %<>% transform(list_label=do.call(paste, c(listLabels, sep="__")))
 
-geneLists %>% inner_join(listLabels) %>%
-    ggplot(aes(list_label)) +
-    geom_bar() +
-    coord_flip() +
-    ggtitle("gene list sizes to be tested for term enrichment") +
-    ylab("")
-
-
-## todo move to diffex commons
-guess_cp_species <- function(ensIds){
-    an_id <-ensIds[1]
-
-    if(str_detect(an_id, "ENSG")){
-        return("human")
-    }else if(str_detect(an_id, "ENSMUSG")){
-        return("mouse")
-    }else if(str_detect(an_id, "ENSDARG")){
-        return("zebrafish")
-    }else if(str_detect(an_id, "FBgn")){
-        return("fly")
-    }else{
-        stop(paste("could not clusterProfiler species name from ", an_id))
-    }
-}
+stopifnot(unlen(count(geneLists)$n) == 1) # make sure that all input lists have the same length (to avoid that just degs are provided as input
+
+count(geneLists) %>% kable()
 
-guess_anno_db <- function(ensIds){
-    an_id <-ensIds[1]
-
-    if(str_detect(an_id, "ENSG")){
-        return("org.Hs.eg.db")
-    }else if(str_detect(an_id, "ENSMUSG")){
-        return("org.Mm.eg.db")
-    }else if(str_detect(an_id, "ENSDARG")){
-        return("org.Dr.eg.db")
-    }else if(str_detect(an_id, "FBgn")){
-        return("org.Dm.eg.db")
-    }else{
-        stop(paste("could not anno db mart from ", an_id))
-    }
-}
 
 #source("http://bioconductor.org/biocLite.R")
 #biocLite("org.Mm.eg.db")