From: preecej <preecej@localhost>
Date: Tue, 12 Aug 2014 23:44:10 +0000 (+0000)
Subject: The generation of the MSU-RAP dictionary now accepts .2, .3, .4, etc.
X-Git-Url: http://gitweb.planteome.org/?a=commitdiff_plain;h=38c32e12e2bf3a3eb96783a863ff7ea9162a46c1;p=old-jaiswallab-svn%2F.git

The generation of the MSU-RAP dictionary now accepts .2, .3, .4, etc.
isoforms if no .1 is present. This will ensure the inclusion of more
orthology mappings.

svn path=/; revision=581
---

diff --git a/Personnel/preecej/python_singletons/incomparanoid.py b/Personnel/preecej/python_singletons/incomparanoid.py
index 1292f6b..75e4856 100755
--- a/Personnel/preecej/python_singletons/incomparanoid.py
+++ b/Personnel/preecej/python_singletons/incomparanoid.py
@@ -40,6 +40,7 @@ def create_dict_uniprot_map(uniprot_substitution_path) :
         loc = cols[0]
         uniprot = cols[1]
         if loc in dict_uniprot_map :
+            # this should not occur; 1-to-many LOC-to-Uniprot relationships not allowed (only the inverse)
             dict_uniprot_map[loc].add(uniprot)
         else :
             dict_uniprot_map[loc] = uniprot
@@ -85,7 +86,7 @@ def create_ens_map(filtering_loci_path, ensembl_input_path, rap_map_path, recip_
     reference loci
     """
     dict_ens_map = {} # local ensembl orthology dict
-    dict_rap_map = {} # local MSU-RAP dict, w/ only filtered canonical LOC loci (orig. ".1")
+    dict_rap_map = {} # local MSU-RAP dict, using only filtered canonical LOC loci (lowest available: ".1" is first preference, ".2" second preference, and so on)
 
     # generate internal MSU-RAP map    
     RAP_MAP = open(rap_map_path)
@@ -95,14 +96,17 @@ def create_ens_map(filtering_loci_path, ensembl_input_path, rap_map_path, recip_
             rap_id = cols[0].upper()
 
             set_loc_ids = set(cols[1].upper().split(","))
-            # select only the locus w/ a .1 suffix, if it exists
+            # select only the first locus, if it exists
             for loc_id in set_loc_ids :
-                curr_canon_locus = re.match('.*\.1', loc_id)
-                if curr_canon_locus :
-                    dict_rap_map[rap_id] = curr_canon_locus.group(0).rstrip("1").rstrip(".")
+                if loc_id != "NONE" :
+                    canonical = loc_id.split(".")
+                    dict_rap_map[rap_id] = canonical[0]
                     break;                    
     RAP_MAP.close()
 
+    #for keys, values in dict_rap_map.items() :
+    #    print(keys + " | " + values)
+
     # generate ref loci filter 
     FILTER = open(filtering_loci_path)
     loci_filter = set()
@@ -401,11 +405,13 @@ if args.uniprot_substitution :
     dict_uniprot_map = create_dict_uniprot_map(args.uniprot_substitution)
 
 # create projection maps
-dict_inp_map = create_inp_map(args.inparanoid_input_path, dict_uniprot_map)
+if (args.inparanoid_input_path) :
+    dict_inp_map = create_inp_map(args.inparanoid_input_path, dict_uniprot_map)
 dict_ens_map = create_ens_map(args.filtering_loci_path, args.ensembl_input_path, args.rap_map_path, args.reciprocal_id, dict_uniprot_map, 1 if args.confidence_high else 0)
 
 # generate stats and output them
-all_venn_data = compare_maps(dict_ens_map, dict_inp_map, args.comparison_file_path, args.ensembl_output_path, args.inparanoid_output_path)
+if (args.inparanoid_input_path) :
+    all_venn_data = compare_maps(dict_ens_map, dict_inp_map, args.comparison_file_path, args.ensembl_output_path, args.inparanoid_output_path)
 
 if args.generate_reactome_output == 'ensembl' :
     write_reactome_files(dict_ens_map, args.reactome_gene_protein_path, args.reactome_projection_path, args.projection_prefix)