Changeset 36

Timestamp:

Aug 11, 2010, 4:07:46 PM (10 years ago)

Author:

j@…

bzr:base-revision:

j@dannynavarro.net-20100809093614-khwr3ax29z5bvx1p

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-20100809114648-ieq7nmfd0ljlvleh

bzr:revno:

36

bzr:revprop:branch-nick:

trunk

bzr:root:

trunk

bzr:timestamp:

2010-08-09 13:46:48.609999895 +0200

bzr:user-agent:

bzr2.1.2+bzr-svn1.0.3

svn:original-date:

2010-08-09T11:46:48.610000Z

Message:

_parse_psms method defined, need to implement functions that get called

File:

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

trunk/mzcms/parsers.py

@@ -10 +10 @@
 from string import Template
 from collections import defaultdict
+from collections import namedtuple
 
 SEPARATOR = '--gc0p4Jq0M2Yt08jU534c0p'
@@ -15 +16 @@
         r'Content-Type: application/x-Mascot; name="$section"')
 
-PEP_REGEX = re.compile(r'^q(\d+)'    # nativeid
-                        '_p\d+'      # rank
-                        '=\d+?'      # missed cleavages
-                        ',[\.\d]+?'  # peptide mass
-                        ',[-\.\d]+?' # delta mass
-                        ',\d+?'      # n ions matches
-                        ',(\w+?)'    # peptide string
-                        ',\d+?'      # peaks used for Ions1
-                        ',(\d+?)'    # modstring
-                        ',[\.\d]+?'  # score
-                        ',\d+?'      # ion series found
-                        ',\d+?'      # peaks used for Ions2
-                        ',\d+?'      # peaks used for Ions3
-                        ';(.+)$'     # protein accessions string
+PEP_REGEX = re.compile(r'^q(\d+)'      # nativeid
+                        '_p(\d+)'      # rank
+                        '=\d+?'        # missed cleavages
+                        ',([\.\d]+?)'  # peptide mass
+                        ',([-\.\d]+?)' # delta mass
+                        ',\d+?'        # n ions matches
+                        ',(\w+?)'      # peptide string
+                        ',\d+?'        # peaks used for Ions1
+                        ',(\d+?)'      # modstring
+                        ',([\.\d]+?)'  # score
+                        ',\d+?'        # ion series found
+                        ',\d+?'        # peaks used for Ions2
+                        ',\d+?'        # peaks used for Ions3
+                        ';(.+)$'       # protein accessions string
                         )
 #fraction_regex = re.compile(r'fraction%3a%20(\d+)')
@@ -83 +84 @@
                 import ipdb; ipdb.set_trace()
                 nativeid = int(match.group(1))
-                peptide_str = match.group(2)
-                mods_str = match.group(3)
-                accs = match.group(4)
-                peptide = apply_mods(peptide_str, mods_str)
-                is_decoy = check_decoy(accs, self.decoy_regex)
-                # Assuming that after a peptide line, always the next line is
-                # the line with flanking Aa
-                # q4_p2_terms=E,K:E,K:E,K:E,K:E,K
-                line = dat_file.next().strip()
-                # (('E', 'K'), ('E', 'K'), ('E', 'K'))
-                flanking_pairs = set(tuple(x.split(','))
-                        for x in line.split("=")[1].split(":"))
-                for flanking_pair in flanking_pairs:
-                    annotation = '.'.join((
-                            flanking_pair[0], peptide, flanking_pair[1]))
-                    annotation = annotation.replace('-', '*')
-                    nativeid_psms[nativeid].append(dict(
-                            Annotation=annotation, IsDecoy=is_decoy))
+                rank = int(match.group(2))
+                pep_mass = float(match.group(3))
+                delta_mass = float(match.group(4))
+                pep_seq = match.group(5)
+                mods_str = match.group(6)
+                score = match.group(7)
+                accs = match.group(8)
+                prot_ids = get_prot_ids(accs)
+                is_target = check_target(accs, self.decoy_regex)
+                if is_target and rank == 1 or rank == 2:
+                    mascot_psm = MascotPsm('mascot_psm',
+                            rank=rank,
+                            pep_mass=pep_mass,
+                            pep_seq=pep_seq,
+                            prot_ids=prot_ids
+                            )
+                    nativeid_psms[nativeid].append(mascot_psm)
             elif line.strip() == SEPARATOR:
                 return nativeid_psms