Updated fasta_ltrretriever_lai

Plant-Food-Research-Open · Feb 29, 2024 · f13a9e5 · f13a9e5
1 parent d714bf0
commit f13a9e5
Show file tree

Hide file tree

Showing 3 changed files with 40 additions and 8 deletions.
diff --git a/modules.json b/modules.json
@@ -81,7 +81,7 @@
                 "pfr": {
                     "fasta_ltrretriever_lai": {
                         "branch": "main",
-                        "git_sha": "60ee50c79f8f868fe0b2e48cba543ad1d8fa89b2",
+                        "git_sha": "f148f5384395618fc706b6e2f059bd1ce037d06c",
                         "installed_by": ["subworkflows"]
                     },
                     "fastq_bwa_mem_samblaster": {

diff --git a/subworkflows/pfr/fasta_ltrretriever_lai/main.nf b/subworkflows/pfr/fasta_ltrretriever_lai/main.nf
@@ -24,7 +24,7 @@ workflow FASTA_LTRRETRIEVER_LAI {
     ch_short_ids_fasta              = ch_fasta
                                     | join(CUSTOM_SHORTENFASTAIDS.out.short_ids_fasta, by:0, remainder:true)
                                     | map { meta, fasta, short_ids_fasta ->
-                                        [ meta, short_ids_fasta ?: fasta ]
+                                        if ( fasta ) { [ meta, short_ids_fasta ?: fasta ] }
                                     }
 
     ch_short_ids_tsv                = CUSTOM_SHORTENFASTAIDS.out.short_ids_tsv
@@ -35,6 +35,12 @@ workflow FASTA_LTRRETRIEVER_LAI {
                                     | map { meta, short_ids_tsv, monoploid_seqs ->
                                         map_monoploid_seqs_to_new_ids(meta, short_ids_tsv, monoploid_seqs)
                                     }
+                                    | collectFile(newLine:true)
+                                    | map { seqs ->
+                                        def id = seqs.name.split('.mapped.monoploid.seqs.txt')[0]
+
+                                        [ [ id: id ], seqs ]
+                                    }
     ch_versions                     = ch_versions.mix(CUSTOM_SHORTENFASTAIDS.out.versions.first())
 
     // MODULE: LTRHARVEST
@@ -133,7 +139,7 @@ def map_monoploid_seqs_to_new_ids(meta, short_ids_tsv, monoploid_seqs) {
     def short_ids_head = short_ids_tsv.text.split('\n')[0]
 
     if (short_ids_head == "IDs have acceptable length and character. No change required.") {
-        return [ meta, monoploid_seqs ]
+        return [ "${meta.id}.mapped.monoploid.seqs.txt" ] + monoploid_seqs.text.split('\n')
     }
 
     def orig_to_new_ids = [:]
@@ -142,15 +148,15 @@ def map_monoploid_seqs_to_new_ids(meta, short_ids_tsv, monoploid_seqs) {
         orig_to_new_ids[original_id] = renamed_id
     }
 
-    def output_file = new File("${meta.id}.mapped.monoploid.seqs.txt")
+    def mapped_ids = []
     monoploid_seqs.text.eachLine { original_id ->
         if (!orig_to_new_ids[original_id]) {
             error "Faild to find $original_id in ${monoploid_seqs}" +
             "The monoploid_seqs file is malformed!"
         }
 
-        output_file.append(orig_to_new_ids[original_id])
+        mapped_ids.add(orig_to_new_ids[original_id])
     }
 
-    return [ meta, output_file.toPath() ]
+    return [ "${meta.id}.mapped.monoploid.seqs.txt" ] + mapped_ids
 }
diff --git a/subworkflows/pfr/fasta_ltrretriever_lai/tests/main.nf.test b/subworkflows/pfr/fasta_ltrretriever_lai/tests/main.nf.test
@@ -109,6 +109,8 @@ nextflow_workflow {
 
     test("actinidia_chinensis-genome_21_fasta_gz-without_lai") {
 
+        options '-stub'
+
         setup {
             run("GUNZIP") {
                 script "../../../../modules/nf-core/gunzip"
@@ -148,11 +150,35 @@ nextflow_workflow {
         then {
             assertAll(
                 { assert workflow.success },
-                { assert file(workflow.out.annotation_gff[0][1]).text.contains('Copia_LTR_retrotransposon') },
-                { assert file(workflow.out.ltrlib[0][1]).text.contains('#LTR/Copia') },
+                { assert workflow.out.annotation_gff != null },
+                { assert workflow.out.ltrlib != null },
                 { assert workflow.out.lai_log == [] },
                 { assert workflow.out.lai_out == [] }
             )
         }
     }
+
+    test("empty_fasta_stub") {
+
+        options '-stub'
+
+        when {
+            workflow {
+                """
+                def monoploid_seqs  = new File('test.mono.seq.txt')
+                monoploid_seqs.write("chr_xxxxxxxxxxxxxxx_1")
+
+                input[0]            = Channel.empty()
+                input[1]            = Channel.of( [ [ id:'test' ], monoploid_seqs.toPath() ] )
+                input[2]            = false
+                """
+            }
+        }
+
+        then {
+            assertAll(
+                { assert workflow.success }
+            )
+        }
+    }
 }