Fix duplicated input in single-end fastp command (#2034)

* Update main.nf

* Update meta.yml

* Re-add logos as not staged in a way that works with MultiQC config files

* Removes duplicate input to single end  FASTP

* Fix md5sums

* Separate interleaved from normal SE command to continue supporting `split_by_lines`

* Simplify save_trimmed_fail

* Update modules/fastp/meta.yml

modules/fastp/main.nf

@@ -26,16 +26,15 @@ process FASTP {
 
     script:
     def args = task.ext.args ?: ''
-    // Added soft-links to original fastqs for consistent naming in MultiQC
     def prefix = task.ext.prefix ?: "${meta.id}"
+    def fail_fastq = save_trimmed_fail && meta.single_end ? "--failed_out ${prefix}.fail.fastq.gz" : save_trimmed_fail && !meta.single_end ? "--unpaired1 ${prefix}_1.fail.fastq.gz --unpaired2 ${prefix}_2.fail.fastq.gz" : ''
+    // Added soft-links to original fastqs for consistent naming in MultiQC
     // Use single ended for interleaved. Add --interleaved_in in config.
-    if (meta.single_end) {
-        def fail_fastq = save_trimmed_fail ? "--failed_out ${prefix}.fail.fastq.gz" : ''
+    if ( task.ext.args?.contains('--interleaved_in') ) {
         """
         [ ! -f  ${prefix}.fastq.gz ] && ln -sf $reads ${prefix}.fastq.gz
-        cat ${prefix}.fastq.gz \\
-        | fastp \\
-            --stdin \\
+
+        fastp \\
             --stdout \\
             --in1 ${prefix}.fastq.gz \\
             --thread $task.cpus \\
@@ -45,13 +44,33 @@ process FASTP {
             $args \\
             2> ${prefix}.fastp.log \\
         | gzip -c > ${prefix}.fastp.fastq.gz
+
+        cat <<-END_VERSIONS > versions.yml
+        "${task.process}":
+            fastp: \$(fastp --version 2>&1 | sed -e "s/fastp //g")
+        END_VERSIONS
+        """
+    } else if (meta.single_end) {
+        """
+        [ ! -f  ${prefix}.fastq.gz ] && ln -sf $reads ${prefix}.fastq.gz
+
+        fastp \\
+            --stdout \\
+            --in1 ${prefix}.fastq.gz \\
+            --out1  ${prefix}.fastp.fastq.gz \\
+            --thread $task.cpus \\
+            --json ${prefix}.fastp.json \\
+            --html ${prefix}.fastp.html \\
+            $fail_fastq \\
+            $args \\
+            2> ${prefix}.fastp.log
+
         cat <<-END_VERSIONS > versions.yml
         "${task.process}":
             fastp: \$(fastp --version 2>&1 | sed -e "s/fastp //g")
         END_VERSIONS
         """
     } else {
-        def fail_fastq  = save_trimmed_fail ? "--unpaired1 ${prefix}_1.fail.fastq.gz --unpaired2 ${prefix}_2.fail.fastq.gz" : ''
         def merge_fastq = save_merged ? "-m --merged_out ${prefix}.merged.fastq.gz" : ''
         """
         [ ! -f  ${prefix}_1.fastq.gz ] && ln -sf ${reads[0]} ${prefix}_1.fastq.gz

modules/fastp/meta.yml

@@ -21,7 +21,8 @@ input:
       type: file
       description: |
         List of input FastQ files of size 1 and 2 for single-end and paired-end data,
-        respectively.
+        respectively. If you wish to run interleaved paired-end data,  supply as single-end data
+        but with `--interleaved_in` in your `modules.conf`'s `ext.args` for the module.
   - save_trimmed_fail:
       type: boolean
       description: Specify true to save files that failed to pass trimming thresholds ending in `*.fail.fastq.gz`

tests/modules/fastp/test.yml

@@ -16,7 +16,7 @@
         - "Q20 bases:</td><td class='col2'>12.922000 K (92.984097%)"
         - "single end (151 cycles)"
     - path: output/fastp/test.fastp.json
-      md5sum: 803a024342be986f76486f6ffea15909
+      md5sum: 2616b6791fd89fb1cc2d16a73b9463b0
     - path: output/fastp/test.fastp.log
       contains:
         - "Q20 bases: 12922(92.9841%)"
@@ -73,7 +73,7 @@
         - "Q20 bases:</td><td class='col2'>25.719000 K (93.033098%)"
         - "paired end (151 cycles + 151 cycles)"
     - path: output/fastp/test.fastp.json
-      md5sum: 5b70f43f33778d278a84b3e9270fa114
+      md5sum: 25ec85e3534f380ca2109c894671f1ed
     - path: output/fastp/test.fastp.log
       contains:
         - "Q20 bases: 12922(92.9841%)"
@@ -102,7 +102,7 @@
         - "Q20 bases:</td><td class='col2'>12.922000 K (92.984097%)"
         - "single end (151 cycles)"
     - path: output/fastp/test.fastp.json
-      md5sum: b647fa752d3fe7956d17429bfe27d72c
+      md5sum: 7e10b0b12fab5cff620fdeb1a32392f8
     - path: output/fastp/test.fastp.log
       contains:
         - "Q20 bases: 12922(92.9841%)"