From 7a604f5b36527fe07a8627c107d00f8d7f40295b Mon Sep 17 00:00:00 2001
From: Alejandra Escobar <ales@ebi.ac.uk>
Date: Thu, 18 Apr 2024 11:16:07 +0100
Subject: [PATCH] Dram fixes (#3)

Fixes in dram and pangenome_db_builder script
---
 bin/pangenomeDB_builder_codon.py | 27 ++++++++-------------------
 modules/local/dram/distill.nf    | 20 ++++++++++++--------
 2 files changed, 20 insertions(+), 27 deletions(-)

diff --git a/bin/pangenomeDB_builder_codon.py b/bin/pangenomeDB_builder_codon.py
index 812b396..5e50b64 100755
--- a/bin/pangenomeDB_builder_codon.py
+++ b/bin/pangenomeDB_builder_codon.py
@@ -51,7 +51,10 @@ def metadata_parser(catalogue_metadata):
 def accessory_writer(reps_clusters, loc_prefix):
     for rep in reps_clusters:
         if len(reps_clusters[rep]) > 1:
-            rep_prefix = rep[:-2]
+            if rep.endswith('.1'):
+                rep_prefix = rep[:-4]
+            else:
+                rep_prefix = rep[:-2]
             pan_loc = (
                 loc_prefix
                 + "/species_catalogue/"
@@ -85,7 +88,10 @@ def accessory_writer(reps_clusters, loc_prefix):
 def annot_writer(reps_clusters, loc_prefix, pfam_desc):
     for rep in reps_clusters:
         core_list, core_mgygs = [], []
-        rep_prefix = rep[:-2]
+        if rep.endswith('.1'):
+            rep_prefix = rep[:-4]
+        else:
+            rep_prefix = rep[:-2]
         rep_loc = (
             loc_prefix + "/species_catalogue/" + rep_prefix + "/" + rep + "/genome/"
         )
@@ -205,23 +211,6 @@ def gff_parser(gff_file):
                     att_l = attr.split(";")
                     gene_id = att_l[0].replace("ID=", "")
                     gff_dict[gene_id] = [contig, start, end, strand]
-                    """
-                    kegg_flag, pfam_flag = 0, 0
-                    for attribute in att_l:
-                        att_key,att_val = attribute.split('=')
-                        if att_key == 'KEGG':
-                            ko = att_val.replace('ko:','')
-                            kegg_flag = 1
-                        if att_key == 'Pfam':
-                            pfam = att_val
-                            pfam_flag = 1
-                    if kegg_flag == 0:
-                        ko = '-'
-                    if pfam_flag == 0:
-                        pfam = '-'
-                    gff_dict[gene_id].append(ko)
-                    gff_dict[gene_id].append(pfam)
-                    """
     return gff_dict
 
 
diff --git a/modules/local/dram/distill.nf b/modules/local/dram/distill.nf
index f7a10e9..405d132 100644
--- a/modules/local/dram/distill.nf
+++ b/modules/local/dram/distill.nf
@@ -14,8 +14,8 @@ process DRAM_DISTILL {
     val(in_type)    //species or community
 
     output:
-    tuple val(meta), path("*_dram*"), emit: destill_out
-    path "versions.yml"              , emit: versions
+    tuple val(meta), path("*_dram*"), emit: destill_out, optional: true
+    path "versions.yml"             , emit: versions
 
     when:
     task.ext.when == null || task.ext.when
@@ -31,12 +31,16 @@ process DRAM_DISTILL {
     fi
 
     cat $dram_summary >> dram_input.txt  
-    DRAM.py \\
-        distill \\
-        -i dram_input.txt  \\
-        -o dram_out
-    mv dram_out/product.html ${prefix}_${tool}_${in_type}_dram.html
-    mv dram_out/product.tsv ${prefix}_${tool}_${in_type}_dram.tsv
+    line_count=\$(wc -l dram_input.txt)
+
+    if [[ \$line_count -gt 1 ]]; then
+        DRAM.py \\
+            distill \\
+            -i dram_input.txt  \\
+            -o dram_out
+        mv dram_out/product.html ${prefix}_${tool}_${in_type}_dram.html
+        mv dram_out/product.tsv ${prefix}_${tool}_${in_type}_dram.tsv
+    fi
 
     cat <<-END_VERSIONS > versions.yml
     "${task.process}":