Now using PFAMs from eggnog if description is '-'

Plant-Food-Research-Open · Apr 30, 2024 · 18dc82c · 18dc82c
1 parent eba68eb
commit 18dc82c
Show file tree

Hide file tree

Showing 2 changed files with 13 additions and 7 deletions.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -3,7 +3,7 @@
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/)
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
-## 0.3.0 - [29-April-2024]
+## 0.3.0 - [30-April-2024]
 
 ### `Added`
 
@@ -48,6 +48,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 39. Now using local tests/stub files for GitHub CI
 40. Now removing iso-forms left by TSEBRA using `AGAT_SPFILTERFEATUREFROMKILLLIST`
 41. Added `pyproject.toml`
+42. Now using PFAMs from eggnog if description is '-'
 
 ### `Fixed`
 

diff --git a/subworkflows/local/gff_store.nf b/subworkflows/local/gff_store.nf
@@ -20,12 +20,17 @@ workflow GFF_STORE {
                                             def cols    = line.split('\t')
                                             def id      = cols[0]
                                             def txt     = cols[7]
+                                            def pfams   = cols[20]
 
-                                            [ id, txt ]
+                                            [ id, txt, pfams ]
+                                        }
+                                        .collect { id, txt, pfams ->
+                                            if ( txt != '-' ) { return [ id, txt ] }
+                                            if ( pfams != '-' ) { return [ id, "PFAMs: $pfams" ] }
+
+                                            [ id, 'No eggnog description and PFAMs' ]
                                         }
-                                        .findAll { id, txt ->
-                                            txt != '-'
-                                        }.collectEntries { id, txt ->
+                                        .collectEntries { id, txt ->
                                             [ id, txt ]
                                         }
 
@@ -57,7 +62,7 @@ workflow GFF_STORE {
 
                                             def anno    = tx_annotations.containsKey(tx_id)
                                                         ? URLEncoder.encode(tx_annotations[tx_id], "UTF-8").replace('+', '%20')
-                                                        : URLEncoder.encode('hypothetical protein | no eggnog hit', "UTF-8").replace('+', '%20')
+                                                        : URLEncoder.encode('Hypothetical protein | no eggnog hit', "UTF-8").replace('+', '%20')
 
                                             gene_tx_annotations[gene_id] += [ ( tx_id ): anno ]
                                         }
@@ -67,7 +72,7 @@ workflow GFF_STORE {
                                             def default_anno = tx_annos.values().first()
 
                                             if ( tx_annos.values().findAll { it != default_anno }.size() > 0 ) {
-                                                return [ gene_id, ( tx_annos + [ 'default': 'differing%20isoform%20descriptions' ] ) ]
+                                                return [ gene_id, ( tx_annos + [ 'default': 'Differing%20isoform%20descriptions' ] ) ]
                                             }
 
                                             [ gene_id, ( tx_annos + [ 'default': default_anno ] ) ]