Mercurial > repos > earlhaminst > ete

changeset 13:ed74587a13c8 draft

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
planemo upload for repository https://github.com/TGAC/earlham-galaxytools/tree/master/tools/ete commit 23b3c7c09a7d391b576e3b19f8b34dd63d636bdc
author earlhaminst
date Thu, 01 Sep 2022 16:11:32 +0000
parents dc32007a6b36
children d40b9a7debe5
files ete_gene_cnv.py ete_homology_classifier.py
diffstat 2 files changed, 7 insertions(+), 7 deletions(-) [+]
line wrap: on
line diff
--- a/ete_gene_cnv.py	Tue Jun 07 08:58:05 2022 +0000
+++ b/ete_gene_cnv.py	Thu Sep 01 16:11:32 2022 +0000
@@ -45,10 +45,10 @@
 
             leaves_parts = [_.split("_") for _ in leaves]
             for i, leaf_parts in enumerate(leaves_parts):
-                if len(leaf_parts) != 2:
+                if len(leaf_parts) < 2:
                     raise Exception("Leaf node '%s' is not in gene_species format" % leaves[i])
 
-            leaves_species = [_[1] for _ in leaves_parts]
+            leaves_species = [_[-1] for _ in leaves_parts]
             species_counter = collections.Counter(leaves_species)
 
             # Assign to ref_species the first element of species_list which
@@ -61,8 +61,8 @@
 
             # Find the gene of the (first) leaf node for the ref_species
             for leaf_parts in leaves_parts:
-                if leaf_parts[1] == ref_species:
-                    species_counter['gene'] = leaf_parts[0]
+                if leaf_parts[-1] == ref_species:
+                    species_counter['gene'] = "_".join(leaf_parts[:-1])
                     break
 
             table.append(species_counter)
--- a/ete_homology_classifier.py	Tue Jun 07 08:58:05 2022 +0000
+++ b/ete_homology_classifier.py	Thu Sep 01 16:11:32 2022 +0000
@@ -29,7 +29,7 @@
     # Genetree nodes are required to be in gene_species format
     leaves_list = [_ for _ in leaves_list if '_' in _]
 
-    species_list = [_.split("_")[1] for _ in leaves_list]
+    species_list = [_.split("_")[-1] for _ in leaves_list]
 
     species_dict = {}
     for species in species_list:
@@ -51,8 +51,8 @@
         for leaf2 in leaves_list[i + 1:]:
             id1 = leaf1.split(":")[1] if ":" in leaf1 else leaf1
             id2 = leaf2.split(":")[1] if ":" in leaf2 else leaf2
-            species1 = id1.split("_")[1]
-            species2 = id2.split("_")[1]
+            species1 = id1.split("_")[-1]
+            species2 = id2.split("_")[-1]
             if species1 == species2:
                 homology_type = 'paralogs'
             else: