Changeset 50

Timestamp:

Aug 11, 2010, 4:11:43 PM (11 years ago)

Author:

j@…

bzr:base-revision:

j@dannynavarro.net-20100810125413-6bt7tb040xha3bsg

bzr:committer:

Danny Navarro <j@dannynavarro.net>

bzr:file-ids:

mzcms/models.py models.py-20100730084238-fjjwldiefr0w07zv-4
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-20100810130148-y2rxfm2tv9v7mbl9

bzr:revno:

50

bzr:revprop:branch-nick:

trunk

bzr:root:

trunk

bzr:timestamp:

2010-08-10 15:01:48.036999941 +0200

bzr:user-agent:

bzr2.1.2+bzr-svn1.0.3

svn:original-date:

2010-08-10T13:01:48.037000Z

Message:

parses entire individual dat files correctly

Location:

trunk/mzcms

Files:

• models.py (modified) (2 diffs)
• parsers.py (modified) (6 diffs)

trunk/mzcms/models.py

@@ -24 +24 @@
         self.native_id = str(native_id)
         self.sequence = str(sequence)
-        self.pep_refs = tuple(pep_refs)
+        self.pep_refs = pep_refs
 
 class Peptides(Folder):
@@ -60 +60 @@
 class Psm(Persistent):
     """A Peptide to Spectrum Match"""
-    def __init__(self, score, rank, delta_mass, delta_score,
-                 sequence_rank2, pep_ref, spec_ref):
+    def __init__(self, score, rank, delta_mass, pep_ref, spec_ref):
         self.score = float(score)
         self.rank = int(rank)
         self.delta_mass = float(delta_mass)
-        self.delta_score = float(delta_score)
-        self.sequence_rank2 = str(sequence_rank2)
-        self.pep_ref = str(pep_ref)
-        self.spec_ref = str(spec_ref)
+        #self.delta_score = float(delta_score)
+        #self.sequence_rank2 = str(sequence_rank2)
+        self.pep_ref = pep_ref
+        self.spec_ref = spec_ref
 
 def appmaker(zodb_root):

trunk/mzcms/parsers.py

@@ -53 +53 @@
                  scan_str=r'FinneganScanNumber%3a%20(\d+)%20',
                  rawfn_str=r'RawFile%3a%20(.+raw)',
-                 decoy_str=r'^IPI:REV_:IPI',
-                 contaminant_str=r'^IPI:CON_:IPI',
+                 decoy_str=r'^REV_IPI',
+                 contaminant_str=r'^CON_IPI',
                  # XXX: Fix default factories
                  ):
@@ -64 +64 @@
         self.rawfn_regex = re.compile(rawfn_str)
         self.protein_factory = protein_factory
-        self.peptide_factory = protein_factory
+        self.peptide_factory = peptide_factory
         self.spectrum_factory=spectrum_factory
         self.psm_factory=psm_factory
@@ -131 +131 @@
                 is_target = check_target(prot_ids,
                         self.non_target_regexes)
-                if is_target and rank == 1 or rank == 2:
+                if is_target and (rank == 1 or rank == 2):
                     mascot_psm = MascotPsm(
                             rank=rank,
@@ -173 +173 @@
         nativeid_psms = self._parse_psms(dat_file)
         extraspec = self._parse_extraspec(dat_file)
+        proteins = dict()
         spectra = dict()
+        peptides = dict()
+        psms = dict()
         for summ, extra in izip(summspec, extraspec):
-            spectrum = self.spectrum_factory(prec_mz=summ[0],
-                                             prec_charge=summ[1],
-                                             run=extra[0],
-                                             scan=extra[1],
-                                             peaks="TBI")
+            spectrum = self.spectrum_factory(
+                    prec_mz=summ[0],
+                    prec_charge=summ[1],
+                    run=extra[0],
+                    scan=extra[1],
+                    peaks="TBI"
+                    )
             spectra[(spectrum.scan, spectrum.run)] = spectrum
+        for nativeid, m_psms in nativeid_psms.iteritems():
+            for m_psm in m_psms:
+                peptide = self.peptide_factory(
+                        sequence=m_psm.pep_seq,
+                        mass=m_psm.pep_mass
+                        )
+                if peptide.sequence not in peptides:
+                    peptides[peptide.sequence] = peptide
+                run, spec = extraspec[nativeid - 1]
+                spec_id = spec, run
+                psm = self.psm_factory(
+                        score=m_psm.score,
+                        rank=m_psm.rank,
+                        delta_mass=m_psm.delta_mass,
+                        pep_ref=peptide,
+                        spec_ref=spectra[spec_id]
+                        )
+                psms[(peptide.sequence, spec_id)] = psm
+                for prot_id in m_psm.prot_ids:
+                    if prot_id not in proteins:
+                        protein = self.protein_factory(
+                                native_id=prot_id,
+                                sequence="TBI",
+                                pep_refs=[peptide]
+                                )
+                        proteins[prot_id] = protein
+                    else:
+                        if peptide not in proteins[prot_id].pep_refs:
+                            proteins[prot_id].pep_refs.append(peptide)
         import ipdb; ipdb.set_trace()
         return proteins, peptides, spectra, psms
@@ -197 +231 @@
     for prot_id in prot_ids:
         for regex in regexes:
-            if not re.match(regex, prot_id):
-                return True
+            if re.match(regex, prot_id):
+                return False
+        return True
 
 def apply_mods(peptide_str, mod_str):
@@ -244 +279 @@
             # TODO: exception for not dat files
             with open(os.path.join(root, dat_fn)) as dat_file:
-                dat_proteins, dat_peptides, \
-                dat_spectra, dat_psms = dat_parser.parse(dat_file)
-                proteins.update(dat_proteins)
-                peptides.update(dat_peptides)
-                spectra.update(dat_spectra)
-                psms.update(dat_psms)
+                ser_proteins, ser_peptides, \
+                ser_spectra, ser_psms = dat_parser.parse(dat_file)
+                proteins.update(ser_proteins)
+                peptides.update(ser_peptides)
+                spectra.update(ser_spectra)
+                psms.update(ser_psms)
     return proteins, peptides, spectra, psms