tanu/funpdbe-validator
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

Merging improvements from https://github.com/rdk/funpdbe-validator rdk@github - "more specific information in mismatch messages"

Browse source
This commit is contained in:
Mihaly Varadi 2019-02-20 15:08:30 +00:00
parent b37b155bfc
commit 89ac85e304
2 changed files with 32 additions and 24 deletions
Download patch file Download diff file Expand all files Collapse all files

18
tests/test_residue_index.py
View file

@ -42,7 +42,7 @@ def mock_get_residue_numbering_true(self):
return True
def mock_compare_residue_number(self, foo, bar):
def mock_compare_residue_number(self, foo, bar, asd):
return False
@ -67,16 +67,16 @@ class TestCheckResidueIndices(TestCase):
self.assertIsNone(bad_cri._set_pdb_id())
def test_check_numbering(self):
result = self.cri._check_numbering({}, {})
result = self.cri._check_numbering({}, {}, "A")
self.assertFalse(result)
self.cri._compare_residue_number = mock_compare_residue_number
result = self.cri._check_numbering({}, {"residues": [{"pdb_res_label": 0, "aa_type": "ALA"}]})
result = self.cri._check_numbering({}, {"residues": [{"pdb_res_label": 0, "aa_type": "ALA"}]}, "A")
self.assertFalse(result)
def test_get_residue_numbering(self):
mock_data = {"chain_label": "A"}
self.cri.pdb_id = "1CBS"
self.cri._check_numbering = lambda x, y : True
self.cri._check_numbering = lambda x, y, z : True
result = self.cri._get_residue_numbering(mock_data)
self.assertTrue(result)
self.cri.pdb_id = "2H58"
@ -84,7 +84,7 @@ class TestCheckResidueIndices(TestCase):
self.assertFalse(result)
def test_recursive_loop(self):
result = self.cri._recursive_loop([{"foo": "bar"}], "foo", None, None)
result = self.cri._recursive_loop([{"foo": "bar"}], "foo", None, None, "A")
self.assertFalse(result)
def test_with_bad_numbering(self):
@ -94,16 +94,16 @@ class TestCheckResidueIndices(TestCase):
def test_process_residues(self):
result = self.cri._process_residues(
[{"author_residue_number": 1, "residue_name": "ALA", "author_insertion_code": ""}], "1", "ALA")
[{"author_residue_number": 1, "residue_name": "ALA", "author_insertion_code": ""}], "1", "ALA", "A")
self.assertTrue(result)
result = self.cri._process_residues(
[{"author_residue_number": 1, "residue_name": "ALA", "author_insertion_code": "C"}], "1C", "ALA")
[{"author_residue_number": 1, "residue_name": "ALA", "author_insertion_code": "C"}], "1C", "ALA", "A")
self.assertTrue(result)
result = self.cri._process_residues(
[{"author_residue_number": 2, "residue_name": "ALA", "author_insertion_code": ""}], "1", "ALA")
[{"author_residue_number": 2, "residue_name": "ALA", "author_insertion_code": ""}], "1", "ALA", "A")
self.assertFalse(result)
result = self.cri._process_residues(
[{"author_residue_number": 1, "residue_name": "ALA", "author_insertion_code": ""}], "1", "HIS")
[{"author_residue_number": 1, "residue_name": "ALA", "author_insertion_code": ""}], "1", "HIS", "A")
self.assertFalse(result)
def test_with_multichain(self):

38
validator/residue_index.py
View file

@ -77,9 +77,9 @@ class ResidueIndexes(object):
if not residue_numbering.keys():
self.mismatches.append("No residues in PDB for this entry - probably obsoleted entry")
return False
return self._check_numbering(residue_numbering, chain_data)
return self._check_numbering(residue_numbering, chain_data, chain_id)
def _check_numbering(self, residue_numbering, chain_data):
def _check_numbering(self, residue_numbering, chain_data, chain_id):
"""
This method loops through all the residues in a chain
and call the residue index comparator method
@ -92,11 +92,13 @@ class ResidueIndexes(object):
for residue in chain_data["residues"]:
depositor_residue_number = residue["pdb_res_label"]
depositor_aa_type = residue["aa_type"]
if not self._compare_residue_number(depositor_residue_number, depositor_aa_type, residue_numbering):
if not self._compare_residue_number(depositor_residue_number, depositor_aa_type, residue_numbering,
chain_id):
return False
return True
def _compare_residue_number(self, depositor_residue_number, depositor_aa_type, residue_numbering):
def _compare_residue_number(self, depositor_residue_number, depositor_aa_type, residue_numbering,
depositor_chain_id):
"""
This method starts looping through the substructure of the PDBe API data
:param depositor_residue_number: Residue number provided by the user
@ -105,9 +107,10 @@ class ResidueIndexes(object):
:return: True is residue numbering is valid, False if not
"""
molecules = residue_numbering[self.pdb_id]["molecules"]
return self._recursive_loop(molecules, "chains", depositor_residue_number, depositor_aa_type)
return self._recursive_loop(molecules, "chains", depositor_residue_number, depositor_aa_type,
depositor_chain_id)
def _recursive_loop(self, data, label, depositor_residue_number, depositor_aa_type):
def _recursive_loop(self, data, label, depositor_residue_number, depositor_aa_type, depositor_chain_id):
"""
A recursive loop that goes down to residue level and processes all residues
:param data: JSON data
@ -120,13 +123,14 @@ class ResidueIndexes(object):
for item in data:
sub_data = item[label]
if label == "chains":
flag = self._recursive_loop(sub_data, "residues", depositor_residue_number, depositor_aa_type)
flag = self._recursive_loop(sub_data, "residues", depositor_residue_number, depositor_aa_type,
depositor_chain_id)
elif label == "residues":
return self._process_residues(sub_data, depositor_residue_number, depositor_aa_type)
return self._process_residues(sub_data, depositor_residue_number, depositor_aa_type, depositor_chain_id)
if label == "chains":
return flag
def _process_residues(self, residues, depositor_residue_number, depositor_aa_type):
def _process_residues(self, residues, depositor_residue_number, depositor_aa_type, depositor_chain_id):
"""
This method grabs the residue information and call the comparator if the
residue number of PDBe is the same as the user input
@ -136,12 +140,16 @@ class ResidueIndexes(object):
:return: True is residue numbering is valid, False if not
"""
for residue in residues:
if "%i%s" % (residue["author_residue_number"], residue["author_insertion_code"]) == depositor_residue_number:
return self._make_comparison(residue["residue_name"], depositor_aa_type, depositor_residue_number)
self.mismatches.append("residue numbering is completely mismatched between data and PDB entry")
if "%i%s" % (
residue["author_residue_number"], residue["author_insertion_code"]) == depositor_residue_number:
return self._make_comparison(residue["residue_name"], depositor_aa_type, depositor_residue_number,
depositor_chain_id)
self.mismatches.append(
"residue numbering is completely mismatched between data and PDB entry (invalid residue: %s_%s)" % (
depositor_chain_id, depositor_residue_number))
return False
def _make_comparison(self, residue_name, depositor_aa_type, depositor_residue_number):
def _make_comparison(self, residue_name, depositor_aa_type, depositor_residue_number, depositor_chain_id):
"""
This method does the comparison between two residues that have the same index number
The comparison is between amino acid code
@ -152,7 +160,7 @@ class ResidueIndexes(object):
"""
if residue_name == depositor_aa_type:
return True
mismatch = "residue %s (%s) in data does not match residue %s (%s) in PDB" % (
depositor_residue_number, depositor_aa_type, depositor_residue_number, residue_name)
mismatch = "residue %s_%s (%s) in data does not match residue %s (%s) in PDB" % (
depositor_chain_id, depositor_residue_number, depositor_aa_type, depositor_residue_number, residue_name)
self.mismatches.append(mismatch)
return False