derive annotation from .db packages for affy (nf-core#2825)

* derive annotation from .db packages for affy * Appease eclint * Minor fixes --------- Co-authored-by: Maxime U. Garcia <[email protected]>
fulcrumgenomics · Feb 27, 2023 · e460148 · e460148
1 parent 88810e1
commit e460148
Show file tree

Hide file tree

Showing 5 changed files with 71 additions and 12 deletions.
diff --git a/modules/nf-core/affy/justrma/main.nf b/modules/nf-core/affy/justrma/main.nf
@@ -12,9 +12,10 @@ process AFFY_JUSTRMA {
     tuple val(meta2), path(description)
 
     output:
-    tuple val(meta), path("*.rds"), emit: rds
-    tuple val(meta), path("*.tsv"), emit: tsv
-    path "versions.yml"           , emit: versions
+    tuple val(meta), path("*.rds")                           , emit: rds
+    tuple val(meta), path("matrix.tsv")                      , emit: expression
+    tuple val(meta), path("*.annotation.tsv")                , emit: annotation, optional: true
+    path "versions.yml"                                      , emit: versions
 
     when:
     task.ext.when == null || task.ext.when

diff --git a/modules/nf-core/affy/justrma/meta.yml b/modules/nf-core/affy/justrma/meta.yml
@@ -39,11 +39,15 @@ output:
       description: |
         Groovy Map containing sample information
         e.g. [ id:'test' ]
-  - tsv:
+  - expression:
       type: file
       description: TSV-format intensity matrix
-      pattern: "*.tsv"
-  - rdata:
+      pattern: "matrix.tsv"
+  - annotation:
+      type: file
+      description: TSV-format annotation table
+      pattern: "*.annotation.tsv"
+  - rds:
       type: file
       description: Serialised ExpressionSet object
       pattern: "*.rds"

diff --git a/modules/nf-core/affy/justrma/templates/affy_justrma.R b/modules/nf-core/affy/justrma/templates/affy_justrma.R
@@ -54,15 +54,31 @@ read_delim_flexible <- function(file, header = TRUE, row.names = NULL){
 #' Install the right CDF for a given cel file
 #'
 #' @param celfile A valid path to a CEL file
+#' @param annotation Boolean indication wheter to install the annotation
+#'        package
 #'
 #' @return output The CDF environment or a list detailing the failed locations.
 
-install_cdf <- function(celfile){
+install_cdf_db <- function(celfile, annotation = FALSE){
     library(affyio)
     headdetails <- read.celfile.header(celfile)
     ref.cdfName <- headdetails[[1]]
     cleaned.cdfName <- cleancdfname(ref.cdfName, addcdf = FALSE)
-    cdfFromBioC(paste0(cleaned.cdfName, 'cdf'))
+
+    exts = 'cdf'
+    if (annotation){
+        exts <- c(exts, '.db')
+    }
+
+    for (package in paste0(cleaned.cdfName, exts)){
+        install.packages(
+            package,
+            lib = 'libs',
+            repos = BiocManager::repositories(),
+            dependencies = c("Depends", "Imports")
+        )
+    }
+    cleaned.cdfName
 }
 
 #' Round numeric dataframe columns to fixed decimal places by applying
@@ -116,7 +132,8 @@ opt <- list(
     cdfname = NULL,
     rm.mask = FALSE,
     rm.outliers = FALSE,
-    rm.extra = FALSE
+    rm.extra = FALSE,
+    build_annotation = FALSE
 )
 if (opt\$description == ''){
     opt\$description = NULL
@@ -183,7 +200,7 @@ if (! opt\$file_name_col %in% colnames(sample.sheet)){
 dir.create('libs')
 .libPaths('libs')
 first_cel <- file.path(opt\$celfiles_dir, sample.sheet[[opt\$file_name_col]][1])
-install_cdf(first_cel)
+cdf_name <- install_cdf_db(first_cel, annotation = opt\$build_annotation)
 
 # Run the main function
 
@@ -213,6 +230,41 @@ sampleNames(eset) <- sample.sheet[[opt\$sample_name_col]]
 ################################################
 ################################################
 
+if (opt\$build_annotation){
+
+    # Make some annotation
+
+    dbname <- paste0(cdf_name, '.db')
+    library(dbname, character.only = TRUE)
+    anno <- select(
+        get(dbname),
+        keys=keys(get(dbname), keytype="PROBEID"),
+        columns=c('ENSEMBL', 'ENTREZID', 'SYMBOL', 'GENENAME', 'GENETYPE'),
+        keytype="PROBEID"
+    )
+
+    # Remove duplicates by probe
+    anno <- do.call(
+        rbind,
+        lapply(
+            split(
+                anno,
+                anno\$PROBEID
+            ),
+            function(x) apply(x, 2, function(y) paste(unique(y), collapse=','))
+        )
+    )
+
+    write.table(
+        anno,
+        file = paste0(cdf_name, '.annotation.tsv'),
+        col.names = TRUE,
+        row.names = FALSE,
+        sep = '\t',
+        quote = FALSE
+    )
+}
+
 # R object for other processes to use
 
 saveRDS(eset, file = 'eset.rds')

diff --git a/tests/modules/nf-core/affy/justrma/nextflow.config b/tests/modules/nf-core/affy/justrma/nextflow.config
@@ -3,7 +3,7 @@ process {
     publishDir = { "${params.outdir}/${task.process.tokenize(':')[-1].tokenize('_')[0].toLowerCase()}" }
 
     withName: 'test_affy_justrma:AFFY_JUSTRMA' {
-        ext.args = '--sample_name_col name'
+        ext.args = '--sample_name_col name --build_annotation TRUE'
     }
     withName: UNTAR {
         publishDir = [ enabled: false ]

diff --git a/tests/modules/nf-core/affy/justrma/test.yml b/tests/modules/nf-core/affy/justrma/test.yml
@@ -1,10 +1,12 @@
 - name: affy justrma test_affy_justrma
   command: nextflow run ./tests/modules/nf-core/affy/justrma -entry test_affy_justrma -c ./tests/config/nextflow.config -c ./tests/modules/nf-core/affy/justrma/nextflow.config
   tags:
-    - affy
     - affy/justrma
+    - affy
   files:
     - path: output/affy/eset.rds
+    - path: output/affy/hgu133a2.annotation.tsv
+      md5sum: 3da296cb3b092bf839dda54ebb4fd313
     - path: output/affy/matrix.tsv
       md5sum: 462665542e9531f6d8993b745ce78ae4
     - path: output/affy/versions.yml