nt seqs dict using tripl lookup instead of fasta parser

eharkins · Sep 25, 2018 · 4ac5304 · 4ac5304
1 parent 028d10a
commit 4ac5304
Showing 1 changed file with 5 additions and 10 deletions.
diff --git a/bin/build_olmsted_data.py b/bin/build_olmsted_data.py
@@ -15,7 +15,6 @@ def get_args():
     parser.add_argument('-s', '--sequences-out')
     return parser.parse_args()
 
-
 # Some generic data processing helpers helpers
 
 def comp(f, g):
@@ -74,7 +73,7 @@ def pull_datasets(t):
      "cft.reconstruction:seqmeta": [{"tripl.csv:data": ["bio.seq:id", "cft.seq:cluster_multiplicity", "cft.seq:multiplicity"]}],
      "cft.reconstruction:cluster_aa": [{"bio.seq:set": ["*"]}],
      "cft.reconstruction:asr_tree": ["*"],
-     "cft.reconstruction:asr_seqs": ["*"],
+     "cft.reconstruction:asr_seqs": [{'bio.seq:set': ['bio.seq:id', 'bio.seq:seq']}],
      "cft.reconstruction:cluster": 
      [
       "db:ident",
@@ -157,12 +156,9 @@ def parse_tree_data(s, nt_seqs_dict, aa_seqs_dict, seqmeta_dict):
     t = PhyloTree(s, format=1)
     return create_node_records(t, nt_seqs_dict, aa_seqs_dict, seqmeta_dict)
 
-def parse_fasta_string(fasta_string):
-    return dict([tuple(s.replace('\n', ':', 1).replace('\n','').split(':')) for s in fasta_string.split('>')[1:]])
-
-def create_aa_seqs_dict(aa_seq_records):
+def create_seqs_dict(seq_records):
     d = dict()
-    for record in aa_seq_records:
+    for record in seq_records:
         d[record["bio.seq:id"].pop()] = record["bio.seq:seq"].pop()
     return d
 
@@ -177,10 +173,9 @@ def clean_clonal_family_record(d):
     c = d.copy()
     c['cft.reconstruction:cluster'] = c['cft.reconstruction:cluster'][0]
     #c['cft.reconstruction:cluster']['cft.reconstruction:seqmeta'] = c['cft.reconstruction:seqmeta']
-    aa_seqs_dict = create_aa_seqs_dict(list(c['cft.reconstruction:cluster_aa'] )[0]['bio.seq:set'])
+    aa_seqs_dict = create_seqs_dict(list(c['cft.reconstruction:cluster_aa'])[0]['bio.seq:set'])
+    nt_seqs_dict = create_seqs_dict(list(c['cft.reconstruction:asr_seqs'])[0]['bio.seq:set'])
     seqmeta_dict = create_seqmeta_dict(list(c['cft.reconstruction:seqmeta'])[0]['tripl.csv:data'])
-
-    nt_seqs_dict = parse_fasta_string(list(c['cft.reconstruction:asr_seqs'][0]['tripl.file:contents'])[0])
     if(c['cft.reconstruction:asr_tree'][0].get('tripl.file:contents')):
         c['cft.reconstruction:cluster']['cft.reconstruction:asr_tree'] = parse_tree_data(list(c['cft.reconstruction:asr_tree'][0]['tripl.file:contents'])[0], nt_seqs_dict, aa_seqs_dict, seqmeta_dict)
     c = c['cft.reconstruction:cluster']