Changeset 61

Timestamp:

Aug 11, 2010, 4:14:54 PM (11 years ago)

Author:

j@…

bzr:base-revision:

j@dannynavarro.net-20100811130247-m5vcvd0g77y9h9ko

bzr:committer:

Danny Navarro <j@dannynavarro.net>

bzr:file-ids:

mzcms/parsers.py parsers.py-20100806092910-g1sxvv1o5b9umkof-1

bzr:mapping-version:

v4

bzr:repository-uuid:

724254b2-fbe6-419d-9466-c04ef4c9d29d

bzr:revision-id:

j@dannynavarro.net-20100811131617-5n3z2dsngary27hz

bzr:revno:

61

bzr:revprop:branch-nick:

trunk

bzr:root:

trunk

bzr:timestamp:

2010-08-11 15:16:17.500000000 +0200

bzr:user-agent:

bzr2.1.2+bzr-svn1.0.3

svn:original-date:

2010-08-11T13:16:17.500000Z

Message:

Added fasta parse function

File:

• trunk/mzcms/parsers.py (modified) (3 diffs)

trunk/mzcms/parsers.py

@@ -264 +264 @@
         writer.writerow(psm)
 
+def parse_fasta(fasta_path, proteins):
+    # XXX: Improve parser, don't seek back
+    """Update protein sequence container from fasta file"""
+    fasta_file = open(fasta_path, 'rb')
+    while True:
+        line = fasta_file.readline().decode('utf-8').strip()
+        if line.startswith('>IPI:IPI'):
+            seq_lines = list()
+            prot_id = line.split('|')[0][5:]
+            while True:
+                try:
+                    line = fasta_file.readline().decode('utf-8').strip()
+                except StopIteration:
+                    return
+                if not line.startswith('>'):
+                    seq_lines.append(line)
+                    last_line = fasta_file.tell()
+                else:
+                    sequence = ''.join(seq_lines)
+                    if prot_id in proteins:
+                        #proteins[prot_id].sequence == 'TBI':
+                        proteins[prot_id].sequence = sequence
+                        fasta_file.seek(last_line)
+                    break
+
 # XXX: Use better defaults for containers and factories 
 def parse_dats(dats_dir, proteins_container=dict, peptides_container=dict,
@@ -281 +306 @@
     import transaction
     for root, dirs, files in os.walk(dats_dir):
-        for dat_fn in files:
-            if fnmatch.fnmatch(dat_fn, '*.dat'):
-                with open(os.path.join(root, dat_fn)) as dat_file:
+        for fn in files:
+            if fnmatch.fnmatch(fn, '*.dat'):
+                with open(os.path.join(root, fn)) as dat_file:
                     ser_proteins, ser_peptides, \
                     ser_spectra, ser_psms = dat_parser.parse(dat_file)
@@ -300 +325 @@
                             psms[str(psm_id)] = psm
                     transaction.savepoint()
+            elif fnmatch.fnmatch(fn, '*.fasta'):
+                fasta_path = os.path.join(root, fn)
+    # XXX: Improve this function, has to return something
+    try:
+        parse_fasta(fasta_path, proteins)
+    except UnboundLocalError:
+        print("You must have the fasta database in the data to parse "
+              "directory")
+
 
 def main():