:fasta bed

src/RNASeq.config

@@ -9,6 +9,9 @@ profiles {
       withName: trimming {
         container = "urqt:d62c1f8"
       }
+      withName: fasta_from_bed {
+        container = "bedtools:2.25.0"
+      }
     }
   }
   psmn {
@@ -32,8 +35,17 @@ profiles {
         queue = 'E5-2670deb128A,E5-2670deb128B,E5-2670deb128C,E5-2670deb128D,E5-2670deb128E,E5-2670deb128F'
         penv = 'openmp16'
        }
+       withName: fasta_from_bed {
+        beforeScript = "source /usr/share/lmod/lmod/init/bash; module use ~/privatemodules"
+        module = "BEDtools/2.25.0"
+        executor = "sge"
+        clusterOptions = "-m e -cwd -V"
+        memory = "20GB"
+        time = "12h"
+        queue = 'monointeldeb128'
+      }
     }
   }
-}
+}                                                                                              .
 
 

src/RNASeq.nf

@@ -44,3 +44,47 @@ UrQt --t 20 --m ${task.cpus} --gz \
 """
 }
 
+
+/*
+* bedtools :
+* Imputs : fasta files
+* Imputs : bed files
+* Output : fasta files
+*/
+/*                      fasta extraction                                     */
+
+params.fasta = "$baseDir/data/fasta/*.fasta"
+params.bed = "$baseDir/data/annot/*.bed"
+
+log.info "fasta file : ${params.fasta}"
+log.info "bed file : ${params.bed}"
+
+Channel
+  .fromPath( params.fasta )
+  .ifEmpty { error "Cannot find any fasta files matching: ${params.fasta}" }
+  .set { fasta_files }
+Channel
+  .fromPath( params.bed )
+  .ifEmpty { error "Cannot find any bed files matching: ${params.bed}" }
+  .set { bed_files }
+
+process fasta_from_bed {
+  tag "${bed.baseName}"
+  cpus 4
+  publishDir "results/fasta/", mode: 'copy'
+
+  input:
+  file fasta from fasta_files
+  file bed from bed_files
+
+  output:
+  file "*_extracted.fasta" into fasta_files_extracted
+
+  script:
+"""
+bedtools getfasta -name \
+-fi ${fasta} -bed ${bed} -fo ${bed.baseName}_extracted.fasta
+"""
+}
+
+