Merge pull request #338 from nf-core/centrifuge-fix

Fix CENTRIFUGE header modification when --out-fmt turned on
nf-core · Jul 28, 2023 · fba2d8c · fba2d8c
2 parents 9fce098 + 3437201
commit fba2d8c
Show file tree

Hide file tree

Showing 6 changed files with 21 additions and 12 deletions.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -34,7 +34,8 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 - [#313](https://github.com/nf-core/taxprofiler/pull/304) Fix pipeline not providing error when database sheet does not have a header (♥ to @noah472 for reporting, fix by @jfy133)
 - [#330](https://github.com/nf-core/taxprofiler/pull/330) Added better tagging to allow disambiguation of Kraken2 steps of Kraken2 vs Bracken (♥ to @MajoroMask for requesting, added by @jfy133)
 - [#334](https://github.com/nf-core/taxprofiler/pull/334) Increase the memory of the FALCO process to 4GB (fix by @LilyAnderssonLee)
-- Improved meta map stability for more robust pipeline resuming (fix by @jfy133)
+- [#332](https://github.com/nf-core/taxprofiler/pull/332) Improved meta map stability for more robust pipeline resuming (fix by @jfy133)
+- [#338](https://github.com/nf-core/taxprofiler/pull/338) Fixed wrong file 'out' file going to `centrifuge kreport` module (♥ to @LilyAnderssonLee for reporting, fix by @jfy133)
 
 ### `Dependencies`
 
@@ -46,6 +47,8 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ### `Deprecated`
 
+- [#338](https://github.com/nf-core/taxprofiler/pull/338) Updated Centrifuge module to not generate (undocumented) SAM alignments by default if --save_centrifuge_reads supplied, to due to a Centrifuge bug modifying profile header. SAM alignments can still be generated if `--out-fmt` supplied in `database.csv` (♥ to @LilyAnderssonLee for reporting, fix by @jfy133)
+
 ## v1.0.1 - Dodgy Dachshund Patch [2023-05-15]
 
 ### `Added`

diff --git a/conf/modules.config b/conf/modules.config
@@ -559,7 +559,7 @@ process {
         publishDir = [
             path: { "${params.outdir}/centrifuge/${meta.db_name}/" },
             mode: params.publish_dir_mode,
-            pattern: '*.{txt,sam,gz}'
+            pattern: '*.{txt,sam,tab,gz}'
         ]
     }
 

diff --git a/modules.json b/modules.json
@@ -42,7 +42,7 @@
                     },
                     "centrifuge/centrifuge": {
                         "branch": "master",
-                        "git_sha": "c8e35eb2055c099720a75538d1b8adb3fb5a464c",
+                        "git_sha": "7b0b4276fbad744a69c35d3cbf9d55fdfc9491bd",
                         "installed_by": ["modules"]
                     },
                     "centrifuge/kreport": {

diff --git a/modules/nf-core/centrifuge/centrifuge/main.nf b/modules/nf-core/centrifuge/centrifuge/main.nf
diff --git a/modules/nf-core/centrifuge/centrifuge/meta.yml b/modules/nf-core/centrifuge/centrifuge/meta.yml
diff --git a/subworkflows/local/profiling.nf b/subworkflows/local/profiling.nf
@@ -227,7 +227,7 @@ workflow PROFILING {
                                         db: it[3]
                                 }
 
-        CENTRIFUGE_CENTRIFUGE ( ch_input_for_centrifuge.reads, ch_input_for_centrifuge.db, params.centrifuge_save_reads, params.centrifuge_save_reads, params.centrifuge_save_reads  )
+        CENTRIFUGE_CENTRIFUGE ( ch_input_for_centrifuge.reads, ch_input_for_centrifuge.db, params.centrifuge_save_reads, params.centrifuge_save_reads  )
         ch_versions            = ch_versions.mix( CENTRIFUGE_CENTRIFUGE.out.versions.first() )
         ch_raw_classifications = ch_raw_classifications.mix( CENTRIFUGE_CENTRIFUGE.out.results )
 
@@ -236,7 +236,11 @@ workflow PROFILING {
                                                 .filter { meta, db -> meta.tool == 'centrifuge' }
                                                 .map { meta, db -> [meta.db_name, meta, db] }
 
-        ch_input_for_centrifuge_kreport = CENTRIFUGE_CENTRIFUGE.out.report
+        // We must combine the _results_ file to get correct output - sending the report file will
+        // weirdly still produce valid-looking output, however the numbers are nonsense.
+        // Unfortunately the Centrifuge documentation for this was unclear as to _which_ outfile
+        // goes into it.
+        ch_input_for_centrifuge_kreport = CENTRIFUGE_CENTRIFUGE.out.results
                                             .map { meta, profile -> [meta.db_name, meta, profile] }
                                             .combine(ch_database_for_centrifugekreport, by: 0)
                                             .multiMap {