add JOIN_SPREAD_ATOM_RECORDS flag to PDBReader

git-svn-id: https://dng.biozentrum.unibas.ch/svn/openstructure/trunk@1916 5a81b35b-ba03-0410-adc8-b2c5c5119f08

modules/io/pymod/__init__.py

@@ -20,7 +20,8 @@ from _io import *
 from ost import mol,conop
 
 def LoadPDB(filename, restrict_chains="", no_hetatms=False,
-            fault_tolerant=False, load_multi=False):
+            fault_tolerant=False, load_multi=False, 
+            join_spread_atom_records=False):
   """
   Load PDB file from disk.
   
@@ -35,6 +36,10 @@ def LoadPDB(filename, restrict_chains="", no_hetatms=False,
   
   If load_multi is set to true, a list of entities will be returned instead
   of only the first.
+  
+  If join_spread_atom_records is set to true, atom records belonging to the 
+  same residue are joined, even if they do not appear sequentially in the PDB
+  file.
   """          
   conop_inst=conop.Conopology.Instance()
   builder=conop_inst.GetBuilder("DEFAULT")
@@ -45,6 +50,8 @@ def LoadPDB(filename, restrict_chains="", no_hetatms=False,
     flags=PDB.SKIP_FAULTY_RECORDS
   if no_hetatms:
     flags|=PDB.NO_HETATMS
+  if join_spread_atom_records:
+    flags|=PDB.JOIN_SPREAD_ATOM_RECORDS
   reader.SetFlags(flags)
   if load_multi:
     ent_list=[]

modules/io/pymod/export_pdb_io.cc

@@ -37,6 +37,7 @@ void export_pdb_io()
     .value("NO_HETATMS", PDB::NO_HETATMS)
     .value("SKIP_FAULTY_RECORDS", PDB::SKIP_FAULTY_RECORDS)
     .value("WRITE_MULTIPLE_MODELS", PDB::WRITE_MULTIPLE_MODELS)
+    .value("JOIN_SPREAD_ATOM_RECORDS", PDB::JOIN_SPREAD_ATOM_RECORDS)
   ;
   class_<PDBReader, boost::noncopyable>("PDBReader", init<String>())
     .def("HasNext", &PDBReader::HasNext)

modules/io/src/mol/pdb_io.hh

@@ -35,6 +35,29 @@ struct PDB {
     WRITE_MULTIPLE_MODELS=4,
     /// \brief enable for PQR
     PQR_FORMAT=8,
+    /// \brief Join atom records into one residue, even if the atom records
+    ///     are not sequential.
+    /// 
+    /// This is useful in the following case:
+    /// 
+    /// \verbatim
+    /// ATOM     43  N  AALA P   4
+    /// ATOM     45  CA AALA P   4 
+    /// ATOM     47  C  AALA P   4
+    /// ATOM     48  O  AALA P   4
+    /// ATOM     49  N  APRO P   5
+    /// ATOM     50  CD APRO P   5
+    /// ATOM     53  CA APRO P   5
+    /// ATOM     55  CB APRO P   5
+    /// ATOM     58  CG APRO P   5
+    /// ATOM     61  C  APRO P   5
+    /// ATOM     62  O  APRO P   5
+    /// ATOM    550  CB AALA P   4
+    /// \endverbatim
+    /// 
+    /// By default, the atom 550 will start a new residue instead of being
+    /// joined with atoms 43-48 into one residue.
+    JOIN_SPREAD_ATOM_RECORDS=16
   } Type; 
 };
 

modules/io/src/mol/pdb_reader.cc

@@ -446,14 +446,21 @@ void PDBReader::ParseAndAddAtom(const String& line, int line_num,
       ++chain_count_;
     }
   }
-
   if(update_residue) {
-    LOGN_DUMP("new residue " << rkey << " " << rnum);
-    curr_residue_=editor.AppendResidue(curr_chain_, rkey, rnum);
+    if (flags_ & PDB::JOIN_SPREAD_ATOM_RECORDS) {
+      curr_residue_=curr_chain_.FindResidue(rnum);
+      if (!curr_residue_.IsValid()) {
+        LOGN_DUMP("new residue " << rkey << " " << rnum);      
+        curr_residue_=editor.AppendResidue(curr_chain_, rkey, rnum);
+        ++residue_count_; 
+      }
+    } else {
+      LOGN_DUMP("new residue " << rkey << " " << rnum);      
+      curr_residue_=editor.AppendResidue(curr_chain_, rkey, rnum);
+      ++residue_count_;      
+    }
     assert(curr_residue_.IsValid());
-    ++residue_count_;
   }
-
   // finally add atom
   LOGN_DUMP("adding atom " << aname << " (" << s_ele << ") @" << apos);
   mol::AtomProp aprop;