fix: CanonizerUtil with NOSTEREO yielded to a wrong result (#165)

__tests__/canonizer.test.js

@@ -0,0 +1,62 @@
+'use strict';
+
+const OCL = require('../core');
+
+test('canonizer chiral', () => {
+  const molecule = OCL.Molecule.fromSmiles('C[C@H](Cl)CC');
+  const idCode = molecule.getIDCode();
+
+  expect(idCode).toBe('gJPHADILuTb@');
+  expect(OCL.CanonizerUtil.getIDCode(molecule, OCL.CanonizerUtil.NORMAL)).toBe(
+    idCode,
+  );
+
+  expect(
+    OCL.CanonizerUtil.getIDCode(molecule, OCL.CanonizerUtil.NOSTEREO),
+  ).toBe('gJPHADILuP@');
+
+  expect(
+    OCL.CanonizerUtil.getIDCode(molecule, OCL.CanonizerUtil.BACKBONE),
+  ).toBe('gJPHADILuP@');
+
+  expect(
+    OCL.CanonizerUtil.getIDCode(molecule, OCL.CanonizerUtil.TAUTOMER),
+  ).toBe('gJPHADILuTb@');
+
+  expect(
+    OCL.CanonizerUtil.getIDCode(molecule, OCL.CanonizerUtil.NOSTEREO_TAUTOMER),
+  ).toBe('gJPHADILuP@');
+
+  molecule.stripStereoInformation();
+  expect(molecule.getIDCode()).toBe('gJPHADILuP@');
+});
+
+test('canonizer tautomer', () => {
+  const molecule = OCL.Molecule.fromSmiles('CC=C(O)CC');
+
+  const idCode = molecule.getIDCode();
+
+  expect(idCode).toBe('gGQ@@drsT@@');
+  expect(OCL.CanonizerUtil.getIDCode(molecule, OCL.CanonizerUtil.NORMAL)).toBe(
+    idCode,
+  );
+
+  expect(
+    OCL.CanonizerUtil.getIDCode(molecule, OCL.CanonizerUtil.NOSTEREO),
+  ).toBe('gGQ@@drsT@@');
+
+  expect(
+    OCL.CanonizerUtil.getIDCode(molecule, OCL.CanonizerUtil.BACKBONE),
+  ).toBe('gGQ@@druT@@');
+
+  expect(
+    OCL.CanonizerUtil.getIDCode(molecule, OCL.CanonizerUtil.TAUTOMER),
+  ).toBe('gGQ@@druTAkabXVOtfBicxX~F@');
+
+  expect(
+    OCL.CanonizerUtil.getIDCode(molecule, OCL.CanonizerUtil.NOSTEREO_TAUTOMER),
+  ).toBe('gGQ@@druTAkabXVOtfBicxX~F@');
+
+  molecule.stripStereoInformation();
+  expect(molecule.getIDCode()).toBe('gGQ@@drsT@@');
+});

scripts/openchemlib/classes.js

@@ -515,11 +515,6 @@ function changeBaseConformer(code) {
     'mFrequency[bondIndex] = rotatableBond[bondIndex].getDefaultFrequencies().clone();',
     'short[] frequencies = rotatableBond[bondIndex].getDefaultFrequencies();\n			mFrequency[bondIndex] = Arrays.copyOf(frequencies, frequencies.length);',
   );
-  code = replaceChecked(
-    code,
-    '\n			mTorsion[bondIndex] = rotatableBond[bondIndex].getDefaultTorsions().clone();',
-    '',
-  );
   return code;
 }
 

src/com/actelion/research/gwt/chemlib/com/Test.java

@@ -0,0 +1,35 @@
+// to run this code: bash buildOpenChemLib && java -classpath build com.Test
+
+package com;
+
+import java.io.IOException;
+import java.util.TreeMap;
+
+import com.actelion.research.chem.Canonizer;
+import com.actelion.research.chem.CanonizerUtil;
+import com.actelion.research.chem.SmilesParser;
+import com.actelion.research.chem.StereoMolecule;
+import com.actelion.research.chem.contrib.HydrogenHandler;
+import com.actelion.research.chem.contrib.DiastereotopicAtomID;
+import com.actelion.research.chem.io.SDFileParser;
+import java.io.File;
+
+
+public class Test {
+
+  public static void main(String[] args) throws Exception {
+
+    SmilesParser sp = new SmilesParser();
+    StereoMolecule mol = new StereoMolecule();
+    sp.parse(mol, "CC");
+
+    System.out.println(mol.getIDCode());
+    mol.stripStereoInformation();
+    System.out.println(mol.getIDCode());
+    System.out.println(CanonizerUtil.getIDCode(mol, CanonizerUtil.IDCODE_TYPE.NORMAL, false));
+    System.out.println(CanonizerUtil.getIDCode(mol, CanonizerUtil.IDCODE_TYPE.NOSTEREO, false));
+    System.out
+        .println(CanonizerUtil.getIDCode(mol, CanonizerUtil.IDCODE_TYPE.NOSTEREO_TAUTOMER, false));
+
+  }
+}

src/com/actelion/research/gwt/chemlib/com/actelion/research/chem/Canonizer.java

@@ -284,141 +284,10 @@ private void canFindNitrogenQualifyingForParity() {
 						continue;
 						}
 
-					if (mMol.getAtomRingBondCount(atom) != 3)
-						continue;
-
-					// For any neutral nitrogen with three ring bonds
-					// we find the smallest ring. If this is not larger than 7 members
-					// then we find that connBond, which does not belong to that ring.
-					// The we find the bridge size (atom count) to where it touches the smallest ring.
-					// We also find the path length from the touch point on the smallest ring back
-					// to the nitrogen atom.
-					int smallRingSize = mMol.getAtomRingSize(atom);
-					if (smallRingSize > 7)
-						continue;
-
-					RingCollection ringSet = mMol.getRingSet();
-					int smallRingNo = 0;
-					while (smallRingNo<ringSet.getSize()) {
-						if (ringSet.getRingSize(smallRingNo) == smallRingSize
-						 && ringSet.isAtomMember(smallRingNo, atom))
-							break;
-
-						smallRingNo++;
-						}
-
-					if (smallRingNo >= RingCollection.MAX_SMALL_RING_COUNT
-					 && smallRingNo == ringSet.getSize())
-						continue;   // very rare case, but found with wrongly highly bridged CSD entry JORFAZ
-
-					int firstBridgeAtom = -1;
-					int firstBridgeBond = -1;
-					for (int i=0; i<3; i++) {
-						int connBond = mMol.getConnBond(atom, i);
-						if (!ringSet.isBondMember(smallRingNo, connBond)) {
-							firstBridgeAtom = mMol.getConnAtom(atom, i);
-							firstBridgeBond = connBond;
-							break;
-							}
-						}
-
-					boolean[] neglectBond = new boolean[mMol.getBonds()];
-					neglectBond[firstBridgeBond] = true;
-					int[] pathAtom = new int[11];
-					int pathLength = mMol.getPath(pathAtom, firstBridgeAtom, atom, 10, neglectBond);
-					if (pathLength == -1)
-						continue;
-
-					int bridgeAtomCount = 1;
-					while (!ringSet.isAtomMember(smallRingNo, pathAtom[bridgeAtomCount]))
-						bridgeAtomCount++;
-
-					int bondCountToBridgeHead = pathLength - bridgeAtomCount;
-
-					int bridgeHead = pathAtom[bridgeAtomCount];
-
-					if (smallRingSize == 6 && bondCountToBridgeHead == 2 && bridgeAtomCount == 3) {
-						if (mMol.getAtomRingBondCount(pathAtom[1]) >= 3) {
-							boolean isAdamantane = false;
-							int[] ringAtom = ringSet.getRingAtoms(smallRingNo);
-							for (int i=0; i<6; i++) {
-								if (atom == ringAtom[i]) {
-									int potentialOtherBridgeHeadIndex = ringSet.validateMemberIndex(smallRingNo,
-											(bridgeHead == ringAtom[ringSet.validateMemberIndex(smallRingNo, i+2)]) ? i-2 : i+2);
-									int potentialOtherBridgeHead = ringAtom[potentialOtherBridgeHeadIndex];
-									if (mMol.getAtomRingBondCount(potentialOtherBridgeHead) >= 3
-									 && mMol.getPathLength(pathAtom[1], potentialOtherBridgeHead, 2, null) == 2)
-										isAdamantane = true;
-									break;
-									}
-								}
-							if (isAdamantane) {
-								mNitrogenQualifiesForParity[atom] = true;
-								continue;
-								}
-							}
-						}
-
-					boolean bridgeHeadIsFlat = (mMol.getAtomPi(bridgeHead) == 1
-											 || mMol.isAromaticAtom(bridgeHead)
-											 || mMol.isFlatNitrogen(bridgeHead));
-					boolean bridgeHeadMayInvert = !bridgeHeadIsFlat
-												&& mMol.getAtomicNo(bridgeHead) == 7
-												&& mMol.getAtomCharge(bridgeHead) != 1;
-
-					if (bondCountToBridgeHead == 1) {
-						if (!bridgeHeadIsFlat && !bridgeHeadMayInvert && smallRingSize <= 4 && bridgeAtomCount <= 3)
-							mNitrogenQualifiesForParity[atom] = true;
+					if (mMol.isPyramidalBridgeHead(atom)) {
+						mNitrogenQualifiesForParity[atom] = true;
 						continue;
 						}
-
-					switch (smallRingSize) {
-					// case 3 is fully handled
-					case 4:	// must be bondCountToBridgeHead == 2
-						if (!bridgeHeadIsFlat && !bridgeHeadMayInvert) {
-							if (bridgeAtomCount <= 4)
-								mNitrogenQualifiesForParity[atom] = true;
-							}
-						break;
-					case 5:	// must be bondCountToBridgeHead == 2
-						if (bridgeHeadMayInvert) {
-							if (bridgeAtomCount <= 3)
-								mNitrogenQualifiesForParity[atom] = true;
-							}
-						else if (!bridgeHeadIsFlat) {
-							if (bridgeAtomCount <= 4)
-								mNitrogenQualifiesForParity[atom] = true;
-							}
-						break;
-					case 6:
-						if (bondCountToBridgeHead == 2) {
-							if (bridgeHeadIsFlat) {
-								if (bridgeAtomCount <= 4)
-									mNitrogenQualifiesForParity[atom] = true;
-								}
-							else if (!bridgeHeadMayInvert) {
-								if (bridgeAtomCount <= 3)
-									mNitrogenQualifiesForParity[atom] = true;
-								}
-							}
-						else if (bondCountToBridgeHead == 3) {
-							if (bridgeHeadIsFlat) {
-								if (bridgeAtomCount <= 6)
-									mNitrogenQualifiesForParity[atom] = true;
-								}
-							else {
-								if (bridgeAtomCount <= 4)
-									mNitrogenQualifiesForParity[atom] = true;
-								}
-							}
-						break;
-					case 7:
-						if (bondCountToBridgeHead == 3) {
-							if (bridgeAtomCount <= 3)
-								mNitrogenQualifiesForParity[atom] = true;
-							}
-						break;
-						}
 					}
 				}
 			}