Use '_' as gene set name separator

':' caused Nextflow errors. E.g. 'arath:wg*' as output file to publishDir
would try to look for an 'arath' file.

cedalion/scripts/gene_sets.py

@@ -40,7 +40,7 @@ import attr
     is_flag=True,
     help=join_multiline('''\
         Output a gene set per species containing all genes of the species. The
-        sets are named `{species}:wg`.
+        sets are named `{species}_wg`.
         '''
     )
 )
@@ -59,7 +59,7 @@ import attr
     type=click.Path(exists=True, dir_okay=False),
     help=join_multiline('''\
         Output a gene set per species per GO term containing all genes of the
-        species with that GO term. The sets are named `{species}:{go_term}`. The
+        species with that GO term. The sets are named `{species}_{go_term}`. The
         argument should be a CSV file with header annotating all genes with GO
         terms. It must have at least a 'gene', 'species' and
         'blast2go_go_terms', 'interproscan_go_terms' column. Multiple GO terms
@@ -162,7 +162,7 @@ def whole_genome_sets(genes, desired_species):
         return item[1]
     gene_sets = groupby(species, genes.items())
     for species_, items in gene_sets.items():
-        name = '{}:wg'.format(species_)
+        name = '{}_wg'.format(species_)
         if species_ in desired_species:
             genes_ = tuple(item[0] for item in items)
             yield GeneSet(name, genes_)
@@ -186,7 +186,7 @@ def go_gene_sets(genes, desired_species, go_file):
     # Yield GeneSet for each group
     for species, groups in go_groups.items():
         for go_term, genes_ in groups.items():
-            name = '{}:{}'.format(species, go_term)
+            name = '{}_{}'.format(species, go_term.replace(':', '_'))
             yield GeneSet(name, genes_)
 
 @attr.s(slots=True, frozen=True)

cedalion/tests/data/gene_sets/all_gene_sets.tsv

 Symbol	Gene set name
-A1	A:wg
-A2	A:wg
-B1	B:wg
-B2	B:wg
-C1	C:wg
-C2	C:wg
-C3	C:wg
-C4	C:wg
-A1	A:GO:0000001
-A2	A:GO:0000001
-B1	B:GO:0000001
-B2	B:GO:0000001
-B1	B:GO:0000003
-B2	B:GO:0000003
-C1	C:GO:0000010
-C2	C:GO:0000010
-C3	C:GO:0000010
\ No newline at end of file
+A1	A_wg
+A2	A_wg
+B1	B_wg
+B2	B_wg
+C1	C_wg
+C2	C_wg
+C3	C_wg
+C4	C_wg
+A1	A_GO_0000001
+A2	A_GO_0000001
+B1	B_GO_0000001
+B2	B_GO_0000001
+B1	B_GO_0000003
+B2	B_GO_0000003
+C1	C_GO_0000010
+C2	C_GO_0000010
+C3	C_GO_0000010
\ No newline at end of file

cedalion/tests/test_gene_sets.py

@@ -73,9 +73,9 @@ def test_whole_genome():
             '--wg',
         ),
         expected=(
-            'A:wg',
-            'B:wg',
-            'C:wg',
+            'A_wg',
+            'B_wg',
+            'C_wg',
         )
     )
 
@@ -90,10 +90,10 @@ def test_go():
             '--go', resource_path('func_annot.csv'),
         ),
         expected=(
-            'A:GO:0000001',
-            'B:GO:0000001',
-            'B:GO:0000003',
-            'C:GO:0000010',
+            'A_GO_0000001',
+            'B_GO_0000001',
+            'B_GO_0000003',
+            'C_GO_0000010',
         )
     )
 
@@ -112,11 +112,11 @@ def test_subset():
             '--wg',
         ),
         expected=(
-            'A:GO:0000001',
-            'A:wg',
-            'B:GO:0000001',
-            'B:GO:0000003',
-            'B:wg',
+            'A_GO_0000001',
+            'A_wg',
+            'B_GO_0000001',
+            'B_GO_0000003',
+            'B_wg',
         )
     )
 
@@ -134,7 +134,7 @@ def test_size():
             '--wg',
         ),
         expected=(
-            'C:GO:0000010',
-            'C:wg',
+            'C_GO_0000010',
+            'C_wg',
         )
     )