added the paths of all the copies on the target genome in the objects Feature

Functions.py

@@ -399,22 +399,20 @@ def add_target_genome_paths(feature_id,target_genome_paths):
         [first_seg,last_seg,walk_name]=match
         feature_path=[walk_name]
 
-        # first check if several copies in the walk
-        # [first_segs_list,last_segs_list]=detect_gene_copies(list_seg,walk_name)
-        # copy_number=len(first_segs_list)
-        # # then get first_seg+last_seg pairs to create the paths
-        # if copy_number==1:
-        #     feature_path.append(get_feature_path(target_genome_paths[walk_name],first_seg,last_seg,walk_name))
-        #     feature.segments_list_target.append(feature_path)
-        # else: # several copies.
-        #     # ADAPT TO ALL PAIRS OF SEGS
-        feature_path.append(get_feature_path(target_genome_paths[walk_name],first_seg,last_seg,walk_name))
-        feature.segments_list_target.append(feature_path)
+        # get the first and last segments of all the copies
+        [first_last_segs_list]=detect_gene_copies(list_seg,walk_name,feature_id)
+        copy_number=0
+        for first_seg,last_seg in first_last_segs_list: # get the feature path for all the copies
+            copy_number+=1
+            copy_id="copy_"+str(copy_number) # get the copy that corresponds to this pair of first_seg,last_seg
+            feature_path.append(copy_id)
+            feature_path.append(get_feature_path(target_genome_paths[walk_name],first_seg,last_seg,walk_name,copy_id,feature_id))
+            feature.segments_list_target.append(feature_path)
             
     if len(list_first_last_segs)==0: # the latter steps expect this list to not be empty.
         feature.segments_list_target.append(['',[]])
 
-def detect_gene_copies(list_seg_source,walk_name):
+def detect_gene_copies(list_seg_source,walk_name,feature_id):
 
     # find all copies of all segments from the gene in the target genome (in both orientations)
     index=0
@@ -462,8 +460,9 @@ def detect_gene_copies(list_seg_source,walk_name):
                 for segment in segments_on_target_genome[new_seg_id][walk_name]:
                     if segment[1]==seg_start:
                         copy_id="copy_"+str(copy_number)
-                        segment.append(copy_id)
-                        seg.append(copy_id)
+                        feat_copy=(feature_id,copy_id)
+                        segment.append(feat_copy)
+                        break
             else: # end of the copy
                 copy_number+=1
                 last_segs_list.append(old_seg_id)
@@ -471,15 +470,17 @@ def detect_gene_copies(list_seg_source,walk_name):
                 for segment in segments_on_target_genome[new_seg_id][walk_name]:
                     if segment[1]==seg_start:
                         copy_id="copy_"+str(copy_number)
-                        segment.append(copy_id)
-                        seg.append(copy_id)
+                        feat_copy=(feature_id,copy_id)
+                        segment.append(feat_copy)
+                        break
         else: 
             if (old_strand==new_strand) and (old_index<new_index): # if the index increases and the strand stays the same, it is the same gene copy
                 for segment in segments_on_target_genome[new_seg_id][walk_name]:
                     if segment[1]==seg_start:
                         copy_id="copy_"+str(copy_number)
-                        segment.append(copy_id)
-                        seg.append(copy_id)
+                        feat_copy=(feature_id,copy_id)
+                        segment.append(feat_copy)
+                        break
             else: # end of the copy
                 copy_number+=1
                 last_segs_list.append(old_seg_id)
@@ -487,23 +488,42 @@ def detect_gene_copies(list_seg_source,walk_name):
                 for segment in segments_on_target_genome[new_seg_id][walk_name]:
                     if segment[1]==seg_start:
                         copy_id="copy_"+str(copy_number)
-                        segment.append(copy_id)
-                        seg.append(copy_id)
+                        feat_copy=(feature_id,copy_id)
+                        segment.append(feat_copy)
+                        break
         # if the strand changes, it is possible that it is an inversion inside the gene. treat this case later
         old_strand=new_strand
         old_index=new_index
         old_seg_id=new_seg_id
     last_segs_list.append(old_seg_id)
 
-    return [first_segs_list,last_segs_list] # return two lists, first_segs and last_segs
+    first_last_segs_list=[]
+    index=0
+    for first_seg in first_segs_list:
+        last_seg=last_segs_list[index]
+        pair=(first_seg,last_seg)
+        first_last_segs_list.append(pair)
+        index+=1
+
+    return [first_last_segs_list] # return a list of pairs (first_seg,last_seg)
 
 def sort_seg_info(seg_info):
     return seg_info[2]
 
 # find the feature's path in target genome walk
-def get_feature_path(target_genome_path,first_seg,last_seg,walk_name):
-    first_seg_index=segments_on_target_genome[first_seg][walk_name][-1][4]
-    last_seg_index=segments_on_target_genome[last_seg][walk_name][-1][4]
+def get_feature_path(target_genome_path,first_seg,last_seg,walk_name,copy_id,feature_id):
+    # look for first_seg and last_seg that has the right copy_id for this feature
+    seg_in_walk=segments_on_target_genome[first_seg][walk_name]
+    for seg_occurence in seg_in_walk:
+        for feat_seg in seg_occurence[5:]:
+            if (feat_seg[0]==feature_id) & (feat_seg[1]==copy_id):
+                first_seg_index=seg_occurence[4] # find first_seg_index
+    seg_in_walk=segments_on_target_genome[last_seg][walk_name]
+    for seg_occurence in seg_in_walk:
+        for feat_seg in seg_occurence[5:]:
+            if (feat_seg[0]==feature_id) & (feat_seg[1]==copy_id):
+                last_seg_index=seg_occurence[4] # find last_seg_index
+
     first_index=min(first_seg_index,last_seg_index)
     last_index=max(first_seg_index,last_seg_index)
     feature_path_target_genome=target_genome_path[first_index:last_index+1]