diff --git a/modules/io/doc/mmcif.rst b/modules/io/doc/mmcif.rst
index c53a8b858d090bd3233170164482d40d50cbf1f9..2eafc9c9bd1f8a49c74902504c9df244a9a6b550 100644
--- a/modules/io/doc/mmcif.rst
+++ b/modules/io/doc/mmcif.rst
@@ -798,7 +798,7 @@ of the annotation available.
See :attr:`operationsintervalls`
- .. function:: PDBize(asu, seqres=None, min_polymer_size=10, transformation=False)
+ .. function:: PDBize(asu, seqres=None, min_polymer_size=None, transformation=False, peptide_min_size=10, nucleicacid_min_size=10, saccharide_min_size=10)
Returns the biological assembly (bio unit) for an entity. The new entity
created is well suited to be saved as a PDB file. Therefore the function
@@ -810,7 +810,8 @@ of the annotation available.
- Each polymer gets its own chain, named A-Z 0-9 a-z.
- The description of non-polymer chains will be put into a generic string
property called description on the residue level.
- - Ligands that resemble a polymer but have less than *min_polymer_size*
+ - Ligands that resemble a polymer but have less than *min_polymer_size* /
+ *peptide_min_size* / *nucleicacid_min_size* / *saccharide_min_size*
residues are assigned the same numeric residue number. The residues are
distinguished by insertion code.
- Sometimes bio units exceed the coordinate system storable in a PDB file.
@@ -830,11 +831,21 @@ of the annotation available.
:type seqres: :class:`~ost.seq.SequenceList`
:param min_polymer_size: The minimal number of residues a polymer needs to
get its own chain. Everything below that number will be sorted into the
- ligand chain.
+ ligand chain. Overrides *peptide_min_size*, *nucleicacid_min_size* and
+ *saccharide_min_size* if set to a value different than None.
:type min_polymer_size: int
:param transformation: If set, return the transformation matrix used to
move the bounding box of the bio unit to the lower left corner.
:type transformation: :class:`bool`
+ :param peptide_min_size: Minimal size to get an individual chain for a
+ polypeptide. Is overridden by *min_polymer_size*.
+ :type peptide_min_size: :class:`int`
+ :param nucleicacid_min_size: Minimal size to get an individual chain for a
+ polynucleotide. Is overridden by *min_polymer_size*.
+ :type nucleicacid_min_size: :class:`int`
+ :param saccharide_min_size: Minimal size to get an individual chain for an
+ oligosaccharide or polysaccharide. Is overridden by *min_polymer_size*.
+ :type saccharide_min_size: :class:`int`
.. class:: MMCifInfoStructDetails
diff --git a/modules/io/pymod/__init__.py b/modules/io/pymod/__init__.py
index cf8063d011627d477fd7401d004d7b6664a621f7..a0a29a6a43a98691faf1f1b54650f1fff5a0600a 100644
--- a/modules/io/pymod/__init__.py
+++ b/modules/io/pymod/__init__.py
@@ -358,9 +358,15 @@ def LoadMMCIF(filename, fault_tolerant=None, calpha_only=None, profile='DEFAULT'
# arguement is the usual 'self'.
# documentation for this function was moved to mmcif.rst,
# MMCifInfoBioUnit.PDBize, since this function is not included in SPHINX.
-def _PDBize(biounit, asu, seqres=None, min_polymer_size=10,
- transformation=False):
- pdbizer = mol.alg.PDBize(min_polymer_size=min_polymer_size)
+def _PDBize(biounit, asu, seqres=None, min_polymer_size=None,
+ transformation=False, peptide_min_size=10, nucleicacid_min_size=10,
+ saccharide_min_size=10):
+ if min_polymer_size is not None:
+ pdbizer = mol.alg.PDBize(min_polymer_size=min_polymer_size)
+ else:
+ pdbizer = mol.alg.PDBize(peptide_min_size=peptide_min_size,
+ nucleicacid_min_size=nucleicacid_min_size,
+ saccharide_min_size=saccharide_min_size)
chains = biounit.GetChainList()
c_intvls = biounit.GetChainIntervalList()
diff --git a/modules/mol/alg/pymod/wrap_mol_alg.cc b/modules/mol/alg/pymod/wrap_mol_alg.cc
index fd5b6d94de273dc57317e270d58596534fb5131f..d36877c0fb9024c9a200ca0e327fd9b68f900cbd 100644
--- a/modules/mol/alg/pymod/wrap_mol_alg.cc
+++ b/modules/mol/alg/pymod/wrap_mol_alg.cc
@@ -439,7 +439,10 @@ BOOST_PYTHON_MODULE(_ost_mol_alg)
class_<mol::alg::PDBize>("PDBize",
- init<int>(arg("min_polymer_size")=10))
+ init<int,int,int>((arg("peptide_min_size"),
+ arg("nucleicacid_min_size"),
+ arg("saccharide_min_size"))))
+ .def(init<int>(arg("min_polymer_size")=10))
.def("Add", &mol::alg::PDBize::Add,
(arg("asu"), arg("transformations"), arg("seqres")))
.def("Finish", &mol::alg::PDBize::Finish, arg("shift_to_fit")=true)
diff --git a/modules/mol/alg/src/pdbize.cc b/modules/mol/alg/src/pdbize.cc
index 84853c6f0d20502c94d5bfb38f110ff3abf7b46d..c53749ff4bb06dd589940a1a057536a36b6a7558 100644
--- a/modules/mol/alg/src/pdbize.cc
+++ b/modules/mol/alg/src/pdbize.cc
@@ -76,12 +76,19 @@ void PDBize::Add(EntityView asu, const geom::Mat4List& transforms,
e2 =asu.GetChainList().end(); j != e2; ++j) {
ChainView chain = *j;
int chain_length = chain.GetResidueCount();
- if (chain_length < min_polymer_size_ && seqres.IsValid()) {
+ if (((chain.IsPolypeptide() && chain_length < peptide_min_size_) ||
+ (chain.IsPolynucleotide() && chain_length < nucleicacid_min_size_) ||
+ ((chain.IsOligosaccharide() || chain.IsPolysaccharide()) &&
+ chain_length < saccharide_min_size_)) &&
+ seqres.IsValid()) {
seq::SequenceHandle s = seqres.FindSequence(chain.GetName());
if (s.IsValid())
chain_length = s.GetLength();
}
- if (chain.IsPolymer() && chain_length >= min_polymer_size_) {
+ if ((chain.IsPolypeptide() && chain_length >= peptide_min_size_) ||
+ (chain.IsPolynucleotide() && chain_length >= nucleicacid_min_size_) ||
+ ((chain.IsOligosaccharide() || chain.IsPolysaccharide()) &&
+ chain_length >= saccharide_min_size_)) {
if (*curr_chain_name_ == 0) {
throw std::runtime_error("running out of chain names");
}
diff --git a/modules/mol/alg/src/pdbize.hh b/modules/mol/alg/src/pdbize.hh
index 1ddf66fc5d047d882563b5e0fe7e61dc989f8c86..79ce6744d9baaed6b2ec4dd9d9ad5ee6215929bb 100644
--- a/modules/mol/alg/src/pdbize.hh
+++ b/modules/mol/alg/src/pdbize.hh
@@ -35,7 +35,19 @@ extern const char* WATER_CHAIN_NAME;
class DLLEXPORT_OST_MOL_ALG PDBize {
public:
explicit PDBize(int min_polymer_size=10):
- min_polymer_size_(min_polymer_size), ent_(mol::CreateEntity()),
+ peptide_min_size_(min_polymer_size),
+ nucleicacid_min_size_(min_polymer_size),
+ saccharide_min_size_(min_polymer_size), ent_(mol::CreateEntity()),
+ curr_chain_name_(POLYPEPTIDE_CHAIN_NAMES), needs_adjustment_(false),
+ last_rnum_(0)
+ {}
+
+ explicit PDBize(int peptide_min_size,
+ int nucleicacid_min_size,
+ int saccharide_min_size):
+ peptide_min_size_(peptide_min_size),
+ nucleicacid_min_size_(nucleicacid_min_size),
+ saccharide_min_size_(saccharide_min_size), ent_(mol::CreateEntity()),
curr_chain_name_(POLYPEPTIDE_CHAIN_NAMES), needs_adjustment_(false),
last_rnum_(0)
{}
@@ -45,7 +57,9 @@ public:
EntityHandle Finish(bool shift_to_fit=true);
private:
- int min_polymer_size_;
+ int peptide_min_size_;
+ int nucleicacid_min_size_;
+ int saccharide_min_size_;
EntityHandle ent_;
ChainHandle ligand_chain_;
ChainHandle water_chain_;
diff --git a/modules/mol/alg/tests/test_pdbize.py b/modules/mol/alg/tests/test_pdbize.py
index 5a180b95267938eab7b1b2240ef3a5848fb02fbe..c9f67643a04877d3132de8d5f7a054830a389951 100644
--- a/modules/mol/alg/tests/test_pdbize.py
+++ b/modules/mol/alg/tests/test_pdbize.py
@@ -4,7 +4,6 @@ import os
import random
class TestPDBize(unittest.TestCase):
-
def test_numbers_water_molecules_with_ins_codes(self):
m = mol.CreateEntity()
e = m.EditXCS(mol.BUFFERED_EDIT)
@@ -94,6 +93,187 @@ class TestPDBize(unittest.TestCase):
self.assertEqual(residues[26].number.num, 2)
self.assertEqual(residues[26].number.ins_code, '\0')
+ def _CheckMinSize(self, ost_ent, seq_list, chn_nm_lst, **kwargs):
+ """Check effects of the *_min_size parameter.
+
+ :param ost_ent: OST entity to be PDBized.
+ :type ost_ent: :class:`~ost.mol.EntityHandle`
+ :param seq_list: Sequence list for the chains in ost_ent.
+ :type seq_list: :class:`~ost.seq.SequenceList`
+ :param chn_nm_lst: List of expected chain names in PDBized entity.
+ :type chn_nm_lst: :class:`list` of :class:`str`
+ """
+ if "saccharide_min_size" not in kwargs:
+ kwargs["saccharide_min_size"] = 10
+ if "nucleicacid_min_size" not in kwargs:
+ kwargs["nucleicacid_min_size"] = 10
+ if "peptide_min_size" not in kwargs:
+ kwargs["peptide_min_size"] = 10
+ transformations = geom.Mat4List()
+ transformations.append(geom.Mat4())
+ pdbizer = mol.alg.PDBize(**kwargs)
+ pdbizer.Add(ost_ent.Select(''), transformations, seq_list)
+ pdbized = pdbizer.Finish()
+ self.assertEqual(len(pdbized.chains), len(chn_nm_lst))
+ for i in range(0, len(chn_nm_lst)):
+ self.assertEqual(pdbized.chains[i].name, chn_nm_lst[i])
+ return pdbized
+
+ def test_peptide_min_size(self):
+ """Make sure the peptide_min_size parameter works, place a polypeptide in
+ chain '_'.
+ """
+ m = mol.CreateEntity()
+ e = m.EditXCS(mol.BUFFERED_EDIT)
+ c = e.InsertChain("A");
+ e.SetChainType(c, mol.CHAINTYPE_POLY_PEPTIDE_L)
+ for i in range(10):
+ e.AppendResidue(c, "ALA")
+ seqs = seq.CreateSequenceList()
+ seqs.AddSequence(seq.CreateSequence("LotsOfAlanin", "AAAAAAAAAA"))
+
+ # test that small peptide chains end up in the ligand chain "_"
+ self._CheckMinSize(m, seqs, ["_"], peptide_min_size=11)
+
+ # test again with two small peptide chains
+ c = e.InsertChain("B");
+ e.SetChainType(c, mol.CHAINTYPE_POLY_PEPTIDE_L)
+ for i in range(15):
+ e.AppendResidue(c, "ALA")
+ seqs.AddSequence(seq.CreateSequence("MoreAlanin", "AAAAAAAAAAAAAAA"))
+ self._CheckMinSize(m, seqs, ["_"], peptide_min_size=16)
+
+ # test one peptide in ligand chain, second as polymer chain
+ self._CheckMinSize(m, seqs, ["_", "A"], peptide_min_size=11)
+
+ # actually disabling min. polymer size
+ self._CheckMinSize(m, seqs, ["A", "B"], peptide_min_size=0)
+
+ def test_nucleicacid_min_size(self):
+ """Make sure the nucleicacid_min_size parameter works, place a
+ polynucleotide in chain '_'.
+ """
+ m = mol.CreateEntity()
+ e = m.EditXCS(mol.BUFFERED_EDIT)
+ c = e.InsertChain("A");
+ e.SetChainType(c, mol.CHAINTYPE_POLY_DN)
+ for i in range(10):
+ e.AppendResidue(c, "DA")
+ seqs = seq.CreateSequenceList()
+ seqs.AddSequence(seq.CreateSequence("LotsOfAdenine", "AAAAAAAAAA"))
+
+ # test that small nucleotide chains end up in the ligand chain "_"
+ self._CheckMinSize(m, seqs, ["_"], nucleicacid_min_size=11)
+
+ # test again with two small nucleic acid chains
+ c = e.InsertChain("B");
+ e.SetChainType(c, mol.CHAINTYPE_POLY_DN)
+ for i in range(15):
+ e.AppendResidue(c, "DA")
+ seqs.AddSequence(seq.CreateSequence("MoreAdenine", "AAAAAAAAAAAAAAA"))
+ self._CheckMinSize(m, seqs, ["_"], nucleicacid_min_size=16)
+
+ # test one nucleic acid in ligand chain, second as polymer chain
+ self._CheckMinSize(m, seqs, ["_", "A"], nucleicacid_min_size=11)
+
+ # actually disabling min. polymer size
+ self._CheckMinSize(m, seqs, ["A", "B"], nucleicacid_min_size=0)
+
+ def test_saccharide_min_size(self):
+ """Make sure the saccharide_min_size parameter works, place an
+ oligosaccharide in chain '_'.
+ """
+ m = mol.CreateEntity()
+ e = m.EditXCS(mol.BUFFERED_EDIT)
+ c = e.InsertChain("A");
+ e.SetChainType(c, mol.CHAINTYPE_OLIGOSACCHARIDE)
+ for i in range(10):
+ e.AppendResidue(c, "NAG")
+ seqs = seq.CreateSequenceList()
+
+ # test that small oligosaccharides end up in the ligand chain "_"
+ self._CheckMinSize(m, seqs, ["_"], saccharide_min_size=11)
+
+ # test again with two small oligosaccharide chains
+ c = e.InsertChain("B");
+ e.SetChainType(c, mol.CHAINTYPE_OLIGOSACCHARIDE)
+ for i in range(15):
+ e.AppendResidue(c, "NAG")
+ self._CheckMinSize(m, seqs, ["_"], saccharide_min_size=16)
+
+ # test one oligosaccharide in ligand chain, second as polymer chain
+ self._CheckMinSize(m, seqs, ["_", "A"], saccharide_min_size=11)
+
+ # actually disabling min. polymer size
+
+ def test_peptide_nucleicacid_saccharide_min_sizes(self):
+ """Make sure that all thre thresholds play well together.
+ """
+ m = mol.CreateEntity()
+ e = m.EditXCS(mol.BUFFERED_EDIT)
+ c = e.InsertChain("A");
+ e.SetChainType(c, mol.CHAINTYPE_POLY_PEPTIDE_L)
+ for i in range(10):
+ e.AppendResidue(c, "ALA")
+ seqs = seq.CreateSequenceList()
+ seqs.AddSequence(seq.CreateSequence("LotsOfAlanin", "AAAAAAAAAA"))
+ c = e.InsertChain("B");
+ e.SetChainType(c, mol.CHAINTYPE_POLY_DN)
+ for i in range(10):
+ e.AppendResidue(c, "DA")
+ seqs.AddSequence(seq.CreateSequence("LotsOfAdenine", "AAAAAAAAAA"))
+ c = e.InsertChain("C");
+ e.SetChainType(c, mol.CHAINTYPE_OLIGOSACCHARIDE)
+ for i in range(10):
+ e.AppendResidue(c, "NAG")
+
+ # Check branched entities can be abandoned in the ligand chain while
+ # peptides and nucleic acids live in their own chains.
+ pdbized = self._CheckMinSize(m, seqs, ["A", "B", "_"],
+ saccharide_min_size=11,
+ peptide_min_size=0,
+ nucleicacid_min_size=0)
+ self.assertTrue(pdbized.chains[0].IsPolypeptide())
+ self.assertTrue(pdbized.chains[1].IsPolynucleotide())
+ self.assertEqual(pdbized.chains[2].residues[0].GetStringProp("type"),
+ "oligosaccharide")
+
+ # test to store a short polynucleotide and sugar in the ligand chain but keep
+ # longer polynucleotide and the peptide outside of the ligand chain.
+ c = e.InsertChain("D");
+ e.SetChainType(c, mol.CHAINTYPE_POLY_DN)
+ for i in range(5):
+ e.AppendResidue(c, "DG")
+ seqs.AddSequence(seq.CreateSequence("LotsOfGuanine", "GGGGG"))
+ pdbized = self._CheckMinSize(m, seqs, ["A", "B", "_"],
+ saccharide_min_size=11,
+ peptide_min_size=0,
+ nucleicacid_min_size=6)
+ self.assertTrue(pdbized.chains[0].IsPolypeptide())
+ self.assertTrue(pdbized.chains[1].IsPolynucleotide())
+ self.assertEqual(pdbized.chains[2].residues[0].GetStringProp("type"),
+ "oligosaccharide")
+ self.assertEqual(pdbized.chains[2].residues[-1].GetStringProp("type"),
+ "polydeoxyribonucleotide")
+
+ # test to add a small peptide to the ligand chain
+ c = e.InsertChain("E");
+ e.SetChainType(c, mol.CHAINTYPE_POLY_PEPTIDE_L)
+ for i in range(5):
+ e.AppendResidue(c, "ALA")
+ seqs.AddSequence(seq.CreateSequence("SomeAlanin", "AAAAA"))
+ pdbized = self._CheckMinSize(m, seqs, ["A", "B", "_", "C"],
+ saccharide_min_size=11,
+ peptide_min_size=6,
+ nucleicacid_min_size=3)
+ self.assertTrue(pdbized.chains[0].IsPolypeptide())
+ self.assertTrue(pdbized.chains[1].IsPolynucleotide())
+ self.assertEqual(pdbized.chains[2].residues[0].GetStringProp("type"),
+ "oligosaccharide")
+ self.assertEqual(pdbized.chains[2].residues[-1].GetStringProp("type"),
+ "polypeptide(L)")
+
+
if __name__ == "__main__":
from ost import testutils
testutils.RunTests()