converted r script to a module

main.nf

@@ -88,7 +88,6 @@ workflow {
   //Original scripts used R for wrangling the sintax output, same can be done with a single line of bash code so made the R script an optional module
   if (params.r_processing == true) {
     R_PROCESSING(VSEARCH_SINTAX.out.tsv)
-    ch_versions = ch_versions.mix(R_PROCESSING.out.versions.first())
   }
   else {
     PROCESSING(VSEARCH_SINTAX.out.tsv)

modules/local/r_processing.nf

@@ -7,23 +7,60 @@ process R_PROCESSING {
     input:
     tuple val(meta), path(sintax_tsv)
 
+    //Only process files that have taxonomy predictions 
+    when:
+    sintax_tsv.size() > 0
+
     output:
     tuple val(meta), path('*.classified.tsv')   , emit: fasta
-    path "versions.yml"                      , emit: versions
 
     script:
     """
-    if [ ! -s ${sintax_tsv} ]; then
-        echo "${sintax_tsv} has no sintax predictions" > "${meta.id}.classified.tsv"
-    else
-        Rscript ${projectDir}/bin/summarize_classification_sintax.R --input=${sintax_tsv} --output=${meta.id}
-    fi
-
-    md5sum "${meta.id}.classified.tsv" > "${meta.id}.classified.tsv.md5"
-
-    cat <<-END_VERSIONS > versions.yml
-    "${task.process}":
-        R version: \$(R --version | grep "R version" | sed 's/[(].*//' | sed 's/ //g' | sed 's/[^0-9]*//')
-    END_VERSIONS
+    #!/usr/bin/Rscript
+    library(dplyr)
+
+    data <- read.table("${sintax_tsv}", header=F, sep="\t")
+    data <- data %>% mutate_all(na_if,"")
+
+    id <- data.frame(do.call('rbind', strsplit(as.character(data\$V1), ';', fixed=TRUE)))[1]
+
+    size <- gsub(";", "", data\$V1)
+    size <- data.frame(do.call('rbind', strsplit(size,'=',fixed=TRUE)))[2]
+    size\$X2 <- as.numeric(size\$X2)
+
+    classif <- data.frame(do.call('rbind', strsplit(as.character(data\$V2),',',fixed=TRUE)))
+
+    k <- data.frame(do.call('rbind', strsplit(gsub(")", "",classif\$X1),'(',fixed=TRUE)))
+    k\$X1 <- gsub("k:", "", k\$X1)
+    k\$X2 <- as.numeric(as.character(k\$X2))
+
+    p <- data.frame(do.call('rbind', strsplit(gsub(")", "",classif\$X2),'(',fixed=TRUE)))
+    p\$X1 <- gsub("p:", "", p\$X1)
+    p\$X2 <- as.numeric(as.character(p\$X2))
+
+    c <- data.frame(do.call('rbind', strsplit(gsub(")", "",classif\$X3),'(',fixed=TRUE)))
+    c\$X1 <- gsub("c:", "", c\$X1)
+    c\$X2 <- as.numeric(as.character(c\$X2))
+
+    o <- data.frame(do.call('rbind', strsplit(gsub(")", "",classif\$X4),'(',fixed=TRUE)))
+    o\$X1 <- gsub("o:", "", o\$X1)
+    o\$X2 <- as.numeric(as.character(o\$X2))
+
+    f <- data.frame(do.call('rbind', strsplit(gsub(")", "",classif\$X5),'(',fixed=TRUE)))
+    f\$X1 <- gsub("f:", "", f\$X1)
+    f\$X2 <- as.numeric(as.character(f\$X2))
+
+    g <- data.frame(do.call('rbind', strsplit(gsub(")", "",classif\$X6),'(',fixed=TRUE)))
+    g\$X1 <- gsub("g:", "", g\$X1)
+    g\$X2 <- as.numeric(as.character(g\$X2))
+
+    s <- data.frame(do.call('rbind', strsplit(gsub(")", "",classif\$X7),'(',fixed=TRUE)))
+    s\$X1 <- gsub("s:", "", s\$X1)
+    s\$X2 <- as.numeric(as.character(s\$X2))
+
+    classif <- cbind(id, k, p, c, o, f, g, s, size)
+    colnames(classif)[c(1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16)] <- c("sample", "kingdom", "prob_kingdom", "division", "prob_division", "clade", "prob_clade", "order", "prob_order", "family", "prob_family", "genus", "prob_genus", "species", "prob_species", "size")
+
+    write.table(classif, file=paste("${meta.id}", ".classified.tsv", sep=""), quote=FALSE, sep='\t', row.names = FALSE)
     """
 }