ChemAxon Forum Archive » Storage & search: JChem Base and JChem Oracle Cartridge

parity matching problem

User 870ab5b546

30-03-2010 16:13:40

The target:


<?xml version="1.0" ?>

<cml>

<MDocument>

  <MChemicalStruct>

    <molecule molID="m1">

      <propertyList>

        <property dictRef="Reactor index" title="Reactor index">

          <scalar>1</scalar>

        </property>

      </propertyList>

      <atomArray

          atomID="a1 a2 a3 a4 a5 a6 a7 a8 a9"

          elementType="C C C C C C C O O"

          />

      <bondArray>

        <bond atomRefs2="a7 a1" order="1" />

        <bond atomRefs2="a4 a2" order="1" />

        <bond atomRefs2="a3 a4" order="1" />

        <bond atomRefs2="a4 a5" order="1" />

        <bond atomRefs2="a4 a9" order="1" />

        <bond atomRefs2="a5 a6" order="1" />

        <bond atomRefs2="a6 a7" order="1" />

        <bond atomRefs2="a7 a8" order="1" />

      </bondArray>

    </molecule>

  </MChemicalStruct>

</MDocument>

</cml>

The query:


<?xml version="1.0" ?>

<cml>

<MDocument>

  <MChemicalStruct>

    <molecule molID="m1">

      <propertyList>

        <property dictRef="Reactor index" title="Reactor index">

          <scalar>1</scalar>

        </property>

      </propertyList>

      <atomArray

          atomID="a1 a2 a3 a4 a5 a6 a7 a8 a9"

          elementType="C C C C C C C O O"

          />

      <bondArray>

        <bond atomRefs2="a4 a1" order="1" />

        <bond atomRefs2="a7 a2" order="1" />

        <bond atomRefs2="a3 a4" order="1" />

        <bond atomRefs2="a4 a5" order="1" />

        <bond atomRefs2="a4 a9" order="1" />

        <bond atomRefs2="a5 a6" order="1" />

        <bond atomRefs2="a6 a7" order="1" />

        <bond atomRefs2="a7 a8" order="1" />

      </bondArray>

    </molecule>

  </MChemicalStruct>

</MDocument>

</cml>

The code:


    public static boolean matchExact(Molecule respMol, Molecule authMol,

            int stereoType) throws MolFileException {

        // NOTE: Following options are set using an extensive testing

        // of possible combinations in MolSearch params to get an

        // expected behavior. If changes are made to this, make sure

        // that expected behavior is unchanged for all input combinations

        final String SELF = "MolFunctions.matchExact: ";

        final MolSearch search = new MolSearch();

        final MolSearchOptions searchOpts = search.getSearchOptions();

        searchOpts.setSearchType(SearchConstants.FULL);

        final boolean ignore3D = ((stereoType & IGNORE_TETRAHEDRAL_STEREO) != 0);

        if (ignore3D) {

            debugPrint(SELF + "ignoring 3D stereochemistry.");

            searchOpts.setStereoSearchType(SearchConstants.STEREO_IGNORE);

        } else {

            searchOpts.setStereoSearchType(SearchConstants.STEREO_SPECIFIC);

            final int wavyType = (stereoType & WAVY_AND);

            if (wavyType == WAVY_AND) {

                debugPrint(SELF + "adding WavyBondMatcher.");

                search.addComparator(new WavyBondMatcher());

                // following line needed to avoid bug in JChem 5.1.3_2 and earlier

                searchOpts.setKeepQueryOrder(true);

            } else { // WAVY_XOR

                debugPrint(SELF + "not adding WavyBondMatcher.");

            } // if stereoType == WAVY_AND

        } // if ignore3D

        final boolean ignore2D = ((stereoType & IGNORE_DBL_BOND_STEREO) != 0);

        if (ignore2D) {

            debugPrint(SELF + "ignoring 2D stereochemistry.");

            searchOpts.setDoubleBondStereoMatchingMode(StereoConstants.DBS_NONE);

        } else {

            searchOpts.setDoubleBondStereoMatchingMode(StereoConstants.DBS_ALL);

        } // if ignore2D

        searchOpts.setVagueBondLevel(SearchConstants.VAGUE_BOND_OFF);

                // required for comparing nonaromatized aromatic rings

        searchOpts.setStereoModel(SearchConstants.STEREO_MODEL_GLOBAL);

        searchOpts.setChargeMatching(SearchConstants.CHARGE_MATCHING_EXACT);

        searchOpts.setIsotopeMatching(SearchConstants.ISOTOPE_MATCHING_EXACT);

        searchOpts.setRadicalMatching(SearchConstants.RADICAL_MATCHING_EXACT);

        searchOpts.setValenceMatching(true);

        search.setSearchOptions(searchOpts);

        search.setTarget(respMol);

        search.setQuery(authMol);

        debugPrintMRV(SELF + "Match exact response:\n", respMol);

        debugPrintMRV(SELF + "Match exact author structure:\n", authMol);

        try {

            final boolean match = search.isMatching();

            debugPrint(SELF + "JChem search result is ", match);

            return match;

        } catch (SearchException e2) {

            Utils.alwaysPrint("Error in " + SELF);

            e2.printStackTrace();

            throw new MolFileException(ERROR + e2.getMessage());

        } // try

    } // matchExact(Molecule, Molecule, int)

public final class WavyBondMatcher extends MolComparator

        implements StereoConstants {



    public boolean compareAtoms(int queryAtomNum, int targetAtomNum) {

        boolean match = true;

        final int qAtomNum = getOrigQueryAtom(queryAtomNum);

        final int tAtomNum = getOrigTargetAtom(targetAtomNum);

        if (qAtomNum != -1 && tAtomNum != -1) {

            final int qParity = query.getLocalParity(qAtomNum);

            final int tParity = target.getLocalParity(tAtomNum);

            match = !(qParity == PARITY_EITHER && tParity != PARITY_EITHER);

            debugPrint("WavyBondMatcher.compareAtoms: "

                    + "query = ", query, ", target = ", target,

                    "; \nqAtom ", query.getAtom(qAtomNum), qAtomNum + 1,

                    " has parity ", qParity,

                    "; tAtom ", target.getAtom(tAtomNum), tAtomNum + 1,

                    " has parity ", tParity,

                    " (PARITY_EITHER = ", PARITY_EITHER, "); atoms ",

                    (match ? "match" : "do not match"));

        } // if neither query nor target atom is implicit H

        return match;

    } // compareAtoms(int, int)

                

} // WavyBondMatcher

By all indications, these structures are identical.  However, when I submit them to matchExact(), I am told that they are different.  The problem appears to be in the MolComparator:


WavyBondMatcher.compareAtoms: query = CC(O)CCC(C)(C)O, target = CC(O)CCC(C)(C)O; 

qAtom C1 has parity 0; tAtom C1 has parity 0 (PARITY_EITHER = 3); atoms match

WavyBondMatcher.compareAtoms: query = CC(O)CCC(C)(C)O, target = CC(O)CCC(C)(C)O; 

qAtom C1 has parity 0; tAtom C2 has parity 0 (PARITY_EITHER = 3); atoms match

WavyBondMatcher.compareAtoms: query = CC(O)CCC(C)(C)O, target = CC(O)CCC(C)(C)O; 

qAtom C1 has parity 0; tAtom C3 has parity 0 (PARITY_EITHER = 3); atoms match

WavyBondMatcher.compareAtoms: query = CC(O)CCC(C)(C)O, target = CC(O)CCC(C)(C)O; 

qAtom C1 has parity 0; tAtom C4 has parity 0 (PARITY_EITHER = 3); atoms match

WavyBondMatcher.compareAtoms: query = CC(O)CCC(C)(C)O, target = CC(O)CCC(C)(C)O; 

qAtom C1 has parity 0; tAtom C5 has parity 0 (PARITY_EITHER = 3); atoms match

WavyBondMatcher.compareAtoms: query = CC(O)CCC(C)(C)O, target = CC(O)CCC(C)(C)O; 

qAtom C1 has parity 0; tAtom C6 has parity 0 (PARITY_EITHER = 3); atoms match

WavyBondMatcher.compareAtoms: query = CC(O)CCC(C)(C)O, target = CC(O)CCC(C)(C)O; 

qAtom C1 has parity 0; tAtom C7 has parity 3 (PARITY_EITHER = 3); atoms match

WavyBondMatcher.compareAtoms: query = CC(O)CCC(C)(C)O, target = CC(O)CCC(C)(C)O; 

qAtom C2 has parity 0; tAtom C1 has parity 0 (PARITY_EITHER = 3); atoms match

WavyBondMatcher.compareAtoms: query = CC(O)CCC(C)(C)O, target = CC(O)CCC(C)(C)O; 

qAtom C2 has parity 0; tAtom C2 has parity 0 (PARITY_EITHER = 3); atoms match

WavyBondMatcher.compareAtoms: query = CC(O)CCC(C)(C)O, target = CC(O)CCC(C)(C)O; 

qAtom C2 has parity 0; tAtom C3 has parity 0 (PARITY_EITHER = 3); atoms match

WavyBondMatcher.compareAtoms: query = CC(O)CCC(C)(C)O, target = CC(O)CCC(C)(C)O; 

qAtom C2 has parity 0; tAtom C4 has parity 0 (PARITY_EITHER = 3); atoms match

WavyBondMatcher.compareAtoms: query = CC(O)CCC(C)(C)O, target = CC(O)CCC(C)(C)O; 

qAtom C2 has parity 0; tAtom C5 has parity 0 (PARITY_EITHER = 3); atoms match

WavyBondMatcher.compareAtoms: query = CC(O)CCC(C)(C)O, target = CC(O)CCC(C)(C)O; 

qAtom C2 has parity 0; tAtom C6 has parity 0 (PARITY_EITHER = 3); atoms match

WavyBondMatcher.compareAtoms: query = CC(O)CCC(C)(C)O, target = CC(O)CCC(C)(C)O; 

qAtom C2 has parity 0; tAtom C7 has parity 3 (PARITY_EITHER = 3); atoms match

WavyBondMatcher.compareAtoms: query = CC(O)CCC(C)(C)O, target = CC(O)CCC(C)(C)O; 

qAtom C3 has parity 0; tAtom C1 has parity 0 (PARITY_EITHER = 3); atoms match

WavyBondMatcher.compareAtoms: query = CC(O)CCC(C)(C)O, target = CC(O)CCC(C)(C)O; 

qAtom C3 has parity 0; tAtom C2 has parity 0 (PARITY_EITHER = 3); atoms match

WavyBondMatcher.compareAtoms: query = CC(O)CCC(C)(C)O, target = CC(O)CCC(C)(C)O; 

qAtom C3 has parity 0; tAtom C3 has parity 0 (PARITY_EITHER = 3); atoms match

WavyBondMatcher.compareAtoms: query = CC(O)CCC(C)(C)O, target = CC(O)CCC(C)(C)O; 

qAtom C3 has parity 0; tAtom C4 has parity 0 (PARITY_EITHER = 3); atoms match

WavyBondMatcher.compareAtoms: query = CC(O)CCC(C)(C)O, target = CC(O)CCC(C)(C)O; 

qAtom C3 has parity 0; tAtom C5 has parity 0 (PARITY_EITHER = 3); atoms match

WavyBondMatcher.compareAtoms: query = CC(O)CCC(C)(C)O, target = CC(O)CCC(C)(C)O; 

qAtom C3 has parity 0; tAtom C6 has parity 0 (PARITY_EITHER = 3); atoms match

WavyBondMatcher.compareAtoms: query = CC(O)CCC(C)(C)O, target = CC(O)CCC(C)(C)O; 

qAtom C3 has parity 0; tAtom C7 has parity 3 (PARITY_EITHER = 3); atoms match

WavyBondMatcher.compareAtoms: query = CC(O)CCC(C)(C)O, target = CC(O)CCC(C)(C)O; 

qAtom C4 has parity 3; tAtom C1 has parity 0 (PARITY_EITHER = 3); atoms do not match

WavyBondMatcher.compareAtoms: query = CC(O)CCC(C)(C)O, target = CC(O)CCC(C)(C)O; 

qAtom C4 has parity 3; tAtom C2 has parity 0 (PARITY_EITHER = 3); atoms do not match

WavyBondMatcher.compareAtoms: query = CC(O)CCC(C)(C)O, target = CC(O)CCC(C)(C)O; 

qAtom C4 has parity 3; tAtom C3 has parity 0 (PARITY_EITHER = 3); atoms do not match

WavyBondMatcher.compareAtoms: query = CC(O)CCC(C)(C)O, target = CC(O)CCC(C)(C)O; 

qAtom C4 has parity 3; tAtom C4 has parity 0 (PARITY_EITHER = 3); atoms do not match

WavyBondMatcher.compareAtoms: query = CC(O)CCC(C)(C)O, target = CC(O)CCC(C)(C)O; 

qAtom C4 has parity 3; tAtom C5 has parity 0 (PARITY_EITHER = 3); atoms do not match

WavyBondMatcher.compareAtoms: query = CC(O)CCC(C)(C)O, target = CC(O)CCC(C)(C)O; 

qAtom C4 has parity 3; tAtom C6 has parity 0 (PARITY_EITHER = 3); atoms do not match

WavyBondMatcher.compareAtoms: query = CC(O)CCC(C)(C)O, target = CC(O)CCC(C)(C)O; 

qAtom C4 has parity 3; tAtom C7 has parity 3 (PARITY_EITHER = 3); atoms match

MolFunctions.matchExact: JChem search result is false

Why is it finding a parity of 3 (indicating a wavy bond) for an atom that neither has a wavy bond nor is a stereocenter?


Both molecules are generated by Reactor.  Strangely, in the debugging output, the target is sometimes printed as [#6]C(O)CCC([#6])([#6])O and sometimes as CC(O)CCC(C)(C)O.

User 870ab5b546

31-03-2010 02:49:56

I figured this one out.  The target and query have no coordinates (zero-dimensional).  The local parity calculation is apparently invalid with 0D compounds.  The workaround is to do a 2D clean of the compounds before comparing them.