Merge tag 'openchemlib-2024.10.2'

[maven-release-plugin]  copy for tag openchemlib-2024.10.2

Author: targos

.github/workflows/maven.yml

@@ -11,12 +11,12 @@ jobs:
     strategy:
       matrix:
         # test against latest update of each major Java version:
-        java: [ 8, 11, 17 ]
+        java: [ 8, 11, 17, 21 ]
     name: Java ${{ matrix.java }}
     steps:
-      - uses: actions/checkout@v3
+      - uses: actions/checkout@v4
       - name: Setup java
-        uses: actions/setup-java@v3
+        uses: actions/setup-java@v4
         with:
           distribution: 'zulu'
           java-version: ${{ matrix.java }}

pom.xml

@@ -8,7 +8,7 @@
     Please follow the naming scheme YEAR.MONTH.RELEASE_NO_OF_MONTH
     (eg. 2016.4.1 for second release in Apr 2016)
     -->
-    <version>2024.10.1</version>
+    <version>2024.10.2</version>
 
     <name>OpenChemLib</name>
     <description>Open Source Chemistry Library</description>
@@ -48,6 +48,9 @@
 
     <properties>
         <project.build.sourceEncoding>UTF-8</project.build.sourceEncoding>
+        <maven.compiler.target>1.8</maven.compiler.target>
+        <maven.compiler.source>1.8</maven.compiler.source>
+        <javafx.version>11.0.2</javafx.version>
     </properties>
 
     <!-- Disable doclint for Java 8 -->
@@ -97,17 +100,17 @@
                 <dependency>
                     <groupId>org.openjfx</groupId>
                     <artifactId>javafx-controls</artifactId>
-                    <version>11.0.2</version>
+                    <version>${javafx.version}</version>
                 </dependency>
                 <dependency>
                     <groupId>org.openjfx</groupId>
                     <artifactId>javafx-swing</artifactId>
-                    <version>11.0.2</version>
+                    <version>${javafx.version}</version>
                 </dependency>
                 <dependency>
                     <groupId>org.openjfx</groupId>
                     <artifactId>javafx-web</artifactId>
-                    <version>11.0.2</version>
+                    <version>${javafx.version}</version>
                 </dependency>
             </dependencies>            
         </profile>
@@ -127,12 +130,7 @@
             <plugin>
                 <groupId>org.apache.maven.plugins</groupId>
                 <artifactId>maven-compiler-plugin</artifactId>
-                <version>2.0</version>
-                <configuration>
-                    <source>1.8</source>
-                    <target>1.8</target>
-                    <encoding>UTF-8</encoding>
-                </configuration>
+                <version>3.1</version>
             </plugin>
             <plugin>
                 <groupId>org.apache.maven.plugins</groupId>
@@ -209,7 +207,7 @@
         <connection>scm:git:[email protected]:Actelion/openchemlib.git</connection>
         <developerConnection>scm:git:[email protected]:Actelion/openchemlib.git</developerConnection>
         <url>https://github.com/Actelion/openchemlib</url>
-      <tag>openchemlib-2024.10.1</tag>
+      <tag>openchemlib-2024.10.2</tag>
   </scm>
 
     <distributionManagement>

src/main/java/com/actelion/research/chem/Canonizer.java

@@ -170,12 +170,12 @@ public class Canonizer {
 	private boolean[] mNitrogenQualifiesForParity;
 	private ArrayList<CanonizerFragment> mFragmentList;
 	private ArrayList<int[]> mTHParityNormalizationGroupList;
-	private int mMode,mNoOfRanks,mNoOfPseudoGroups;
+	private final int mMode;
+	private int mNoOfRanks,mNoOfPseudoGroups;
 	private boolean mIsOddParityRound;
-	private boolean mZCoordinatesAvailable,mAllHydrogensAreExplicit;
+	private final boolean mZCoordinatesAvailable,mAllHydrogensAreExplicit;
 	private boolean mCIPParityNoDistinctionProblem;
 	private boolean mEncodeAvoid127;
-
 	private boolean mGraphGenerated;
 	private int mGraphRings,mFeatureBlock;
 	private int[] mGraphAtom;
@@ -184,9 +184,10 @@ public class Canonizer {
 	private int[] mGraphFrom;
 	private int[] mGraphClosure;
 
-	private String		    mIDCode, mEncodedCoords,mMapping;
-	private StringBuilder	mEncodingBuffer;
-	private	int				mEncodingBitsAvail,mEncodingTempData,mAtomBits,mMaxConnAtoms;
+	private String mIDCode, mEncodedCoords,mMapping;
+	private StringBuilder mEncodingBuffer;
+	private	int mEncodingBitsAvail,mEncodingTempData,mMaxConnAtoms;
+	private final int mAtomBits;
 
 	/**
 	 * Runs a canonicalization procedure for the given molecule that creates unique atom ranks,
@@ -204,7 +205,7 @@ public Canonizer(StereoMolecule mol) {
 	 * If mode includes ENCODE_ATOM_CUSTOM_LABELS, than custom atom labels are
 	 * considered for the atom ranking and are encoded into the idcode.<br>
 	 * If mode includes COORDS_ARE_3D, then getEncodedCoordinates() always returns
-	 * a 3D-encoding even if all z-coordinates are 0.0. Otherwise coordinates are
+	 * a 3D-encoding even if all z-coordinates are 0.0. Otherwise, coordinates are
 	 * encoded in 3D only, if at least one of the z-coords is not 0.0.
 	 * @param mol
 	 * @param mode 0 or one or more of CONSIDER...TOPICITY, CREATE..., ENCODE_ATOM_CUSTOM_LABELS, ASSIGN_PARITIES_TO_TETRAHEDRAL_N, COORDS_ARE_3D
@@ -226,17 +227,7 @@ public Canonizer(StereoMolecule mol, int mode) {
 
 		mZCoordinatesAvailable = ((mode & COORDS_ARE_3D) != 0) || mMol.is3D();
 
-		mAllHydrogensAreExplicit = false;
-		if (mMol.getAllAtoms() > mMol.getAtoms()
-		 && !mMol.isFragment()) {
-			mAllHydrogensAreExplicit = true;
-			for (int i=0; i<mMol.getAtoms(); i++) {
-				if (mMol.getImplicitHydrogens(i) != 0) {
-					mAllHydrogensAreExplicit = false;
-					break;
-					}
-				}
-			}
+		mAllHydrogensAreExplicit = (mMol.getImplicitHydrogens() == 0);
 
 		if ((mMode & NEGLECT_ANY_STEREO_INFORMATION) == 0) {
 			mTHParity = new byte[mMol.getAtoms()];
@@ -639,7 +630,7 @@ private boolean canInnerBreakTiesByHeteroTopicity() {
 	private void canBreakTiesRandomly() {
 		for (int atom=0; atom<mMol.getAtoms(); atom++) {
 			mCanBase[atom].init(atom);
-			mCanBase[atom].add(mAtomBits+1, 2*mCanRank[atom]);
+			mCanBase[atom].add(mAtomBits+1, (long)2*mCanRank[atom]);
 			}
 
 		// promote randomly one atom of lowest shared rank.
@@ -895,7 +886,7 @@ private void canRecursivelyFindAllParities() {
 					thParityInfo |= mTHESRGroup[atom];
 					}
 
-				mCanBase[atom].add(2 * parityInfoBits, thParityInfo << parityInfoBits); // generate space for bond parity
+				mCanBase[atom].add(2 * parityInfoBits, (long)thParityInfo << parityInfoBits); // generate space for bond parity
 				}
 
 			for (int bond=0; bond<mMol.getBonds(); bond++) {
@@ -3371,7 +3362,7 @@ public String getEncodedCoordinates() {
 	 * original molecule including coordinates.<br>
 	 * If keepPositionAndScale==false, then coordinate encoding will be relative,
 	 * i.e. scale and absolute positions get lost during the encoding.
-	 * Otherwise the encoding retains scale and absolute positions.<br>
+	 * Otherwise, the encoding retains scale and absolute positions.<br>
 	 * If the molecule has 3D-coordinates and if there are no implicit hydrogen atoms,
 	 * i.e. all hydrogen atoms are explicitly available with their coordinates, then
 	 * hydrogen 3D-coordinates are also encoded despite the fact that the idcode itself does

src/main/java/com/actelion/research/chem/ExtendedMolecule.java

@@ -1445,7 +1445,28 @@ public boolean supportsImplicitHydrogen(int atom) {
 		}
 
 	/**
-	 * Calculates and return the number of implicit hydrogens at atom.
+	 * Calculates and returns the number of implicit hydrogens of the molecule.
+	 * For hydrogens atoms, metals except Al, or a noble gases, 0 is assumed.
+	 * For all other atom kinds the number of implicit hydrogens is basically
+	 * the lowest typical valence that is compatible with the occupied valence,
+	 * minus the occupied valence corrected by atom charge and radical state.
+	 * If this molecule is a fragment, then 0 is returned.
+	 * @return number of implicit hydrogens of the molecule
+	 */
+	public int getImplicitHydrogens() {
+		if (mIsFragment)
+			return 0;
+
+		ensureHelperArrays(cHelperNeighbours);
+		int implicitHydrogens = 0;
+		for (int atom=0; atom<mAtoms; atom++)
+			implicitHydrogens += getImplicitHydrogens(atom);
+
+		return implicitHydrogens;
+		}
+
+	/**
+	 * Calculates and returns the number of implicit hydrogens at atom.
 	 * If atom is itself a hydrogen atom, a metal except Al, or a noble gas,
 	 * then 0 is returned. For all other atom kinds the number of
 	 * implicit hydrogens is basically the lowest typical valence that is compatible
@@ -1463,8 +1484,9 @@ public int getImplicitHydrogens(int atom) {
 		if (!supportsImplicitHydrogen(atom))
 			return 0;
 
-		if ("*".equals(getAtomCustomLabel(atom)))
-			return 0;
+		// attachment points have at least a valence of 1, i.e. they must be connected to something
+		if (mAtomicNo[atom] == 0 || "*".equals(getAtomCustomLabel(atom)))
+			return mAllConnAtoms[atom] == 0 ? 1 : 0;
 
 		ensureHelperArrays(cHelperNeighbours);
 

src/main/java/com/actelion/research/chem/IsomericSmilesCreator.java

@@ -820,7 +820,7 @@ private void appendBondOrderSymbol(int bond, int parentAtom, StringBuilder build
 		if (mEZHalfParity[bond] != 0)
 			builder.append(mEZHalfParity[bond] == 1 ? '/' : '\\');
 		if (mMode == MODE_CREATE_SMARTS) {
-			int bondTypes = mMol.getBondQueryFeatures(Molecule.cBondQFBondTypes | Molecule.cBondQFRareBondTypes);
+			int bondTypes = mMol.getBondQueryFeatures(bond) & (Molecule.cBondQFBondTypes | Molecule.cBondQFRareBondTypes);
 			if (bondTypes != 0) {
 				if ((bondTypes & Molecule.cBondTypeSingle) != 0 && mEZHalfParity[bond] == 0) {
 					builder.append('-');

src/main/java/com/actelion/research/chem/SSSearcher.java

@@ -1322,6 +1322,14 @@ public boolean areBondsSimilar(int moleculeBond, int fragmentBond) {
 			 && ringSize == (mMolecule.getBondQueryFeatures(fragmentBond) & Molecule.cBondQFRingSize) >> Molecule.cBondQFRingSizeShift)
 				return true;
 
+			if (ringSize <= 2) {    // ring size 8-11 is encoded as 1; ring size >=12 is encoded as 2
+				int moleculeRingSize = mMolecule.getBondRingSize(moleculeBond);
+				if (ringSize == 1)
+					return (moleculeRingSize >= 8) && (moleculeRingSize <= 12);
+				else
+					return moleculeRingSize >= 12;
+				}
+
 			boolean found = false;
 			RingCollection ringSet = mMolecule.getRingSet();
 			for (int i=0; i<ringSet.getSize(); i++) {

src/main/java/com/actelion/research/chem/descriptor/DescriptorWeightsHelper.java

@@ -71,7 +71,7 @@ public DescriptorWeightsHelper() {
      * - Charged pp points are set mandatory.
      * @param liSubGraphIndices
      * @param molecule3D
-     * @return
+     * @return array with dimension molecule3D.getAtoms().
      */
     public static int [] calcWeightLabels(List<SubGraphIndices> liSubGraphIndices, Molecule3D molecule3D){
 

src/main/java/com/actelion/research/chem/descriptor/flexophore/MolDistHistViz.java

@@ -41,6 +41,7 @@
 import com.actelion.research.chem.interactionstatistics.InteractionAtomTypeCalculator;
 import com.actelion.research.chem.phesa.pharmacophore.PharmacophoreCalculator;
 import com.actelion.research.chem.phesa.pharmacophore.pp.IPharmacophorePoint;
+import com.actelion.research.util.Formatter;
 import com.actelion.research.util.graph.complete.ICompleteGraph;
 
 import java.io.Serializable;
@@ -1056,7 +1057,40 @@ public String toStringShort(){
 
 		return b.toString();
 	}
-	
+
+	/**
+	 * Shows the weights
+	 * @return
+	 */
+	public String toStringShortWithWeights(){
+
+		if(!finalized)
+			realize();
+		StringBuffer b = new StringBuffer();
+
+		b.append("[");
+		for (int i = 0; i < getNumPPNodes(); i++) {
+			b.append(Formatter.format1(arrWeight[i]));
+			b.append(getNode(i).toStringShort());
+			if(i<getNumPPNodes()-1){
+				b.append(" ");
+			} else {
+				b.append("]");
+			}
+		}
+
+		for (int i = 0; i < getNumPPNodes(); i++) {
+			for (int j = i+1; j < getNumPPNodes(); j++) {
+				byte [] arrHist = getDistHist(i,j);
+
+				if(arrHist!=null)
+					b.append("[" + ArrayUtilsCalc.toString(arrHist) + "]");
+			}
+		}
+
+		return b.toString();
+	}
+
 
 	public boolean equals(Object o) {
 		boolean bEQ=true;

src/main/java/com/actelion/research/chem/descriptor/flexophore/MolDistHistVizHelper.java

@@ -64,6 +64,11 @@ public static MolDistHistViz create(List<PPNodeVizMultCoord> liPPNodeVizMultCoor
         return molDistHistViz;
     }
 
+    /**
+     * Sets the weights in MolDistHistViz. Rule based unification of weight labels for a pharmacophore node.
+     * @param mdhv
+     * @param arrWeightLabel Array dimension must equal number of atoms in the molecule in mdhv.
+     */
     public static void setWeights(MolDistHistViz mdhv, int [] arrWeightLabel){
 
         // The molecule in the descriptor contains the pharmacophore points as additional single atoms.
@@ -75,6 +80,9 @@ public static void setWeights(MolDistHistViz mdhv, int [] arrWeightLabel){
             throw new RuntimeException("Weight vector differs in dimension to number of atoms!");
         }
 
+        //
+        // Rule based unification of weight labels for a pharmacophore node
+        //
         for (PPNodeViz ppNodeViz : mdhv.getNodes()) {
             int [] a = ppNodeViz.getArrayIndexOriginalAtoms();
             int [] w = new int[a.length];
@@ -84,7 +92,7 @@ public static void setWeights(MolDistHistViz mdhv, int [] arrWeightLabel){
 
             int maxWeightLabel = ArrayUtils.max(w);
 
-            // If label is not high, low is king.
+            // If label is not high, low is king. Means the standard wight label is overruled.
             if(maxWeightLabel< DescriptorWeightsHelper.LABEL_WEIGHT_MANDATORY){
                 maxWeightLabel = ArrayUtils.min(w);
             }

src/main/java/com/actelion/research/chem/descriptor/flexophore/completegraphmatcher/HistogramMatchCalculator.java

@@ -57,7 +57,7 @@ public class HistogramMatchCalculator {
 	 * @param indexDistHist2At2
 	 * @return integral of overlap
 	 */
-	public static double getSimilarity(DistHist dh1, int indexDistHist1At1, int indexDistHist1At2, DistHist dh2, int indexDistHist2At1, int indexDistHist2At2){
+	public static double getFractionOverlapIntegral(DistHist dh1, int indexDistHist1At1, int indexDistHist1At2, DistHist dh2, int indexDistHist2At1, int indexDistHist2At2){
 		int indexPostStartDistHist1 = dh1.getIndexPosStartForDistHist(indexDistHist1At1, indexDistHist1At2);
 		int indexPostStartDistHist2 = dh2.getIndexPosStartForDistHist(indexDistHist2At1, indexDistHist2At2);
 		int n = ConstantsFlexophoreGenerator.BINS_HISTOGRAM;
@@ -73,6 +73,26 @@ public static double getSimilarity(DistHist dh1, int indexDistHist1At1, int inde
 		double score = sumMin / sumMax;
 		return score;
 	}
+	public static double getFractionOverlappingBins(DistHist dh1, int indexDistHist1At1, int indexDistHist1At2, DistHist dh2, int indexDistHist2At1, int indexDistHist2At2){
+		int indexPostStartDistHist1 = dh1.getIndexPosStartForDistHist(indexDistHist1At1, indexDistHist1At2);
+		int indexPostStartDistHist2 = dh2.getIndexPosStartForDistHist(indexDistHist2At1, indexDistHist2At2);
+		int n = ConstantsFlexophoreGenerator.BINS_HISTOGRAM;
+		double sumMin = 0;
+		double sumMax = 0;
+
+		double sum1=0;
+		double sum2=0;
+		for (int i = 0; i < n; i++) {
+			int v1 = (dh1.getValueAtAbsolutePosition(indexPostStartDistHist1+i)==0)?0:1;
+			int v2 = (dh2.getValueAtAbsolutePosition(indexPostStartDistHist2+i)==0)?0:1;
+			sum1 += v1;
+			sum2 += v2;
+			sumMin += Math.min(v1, v2);
+			sumMax += Math.max(v1, v2);
+		}
+		double score = sumMin / Math.min(sum1, sum2);
+		return score;
+	}
 
 	public static double getPercentageOverlap(DistHist dh1, int indexDistHist1At1, int indexDistHist1At2, DistHist dh2, int indexDistHist2At1, int indexDistHist2At2){
 		int indexPostStartDistHist1 = dh1.getIndexPosStartForDistHist(indexDistHist1At1, indexDistHist1At2);

src/main/java/com/actelion/research/chem/descriptor/flexophore/completegraphmatcher/ObjectiveBlurFlexophoreHardMatchUncovered.java

@@ -1156,7 +1156,7 @@ public float getSimilarityHistogram(int indexNode1Query, int indexNode2Query, in
 		if(arrSimilarityHistograms[indexHistogramQuery][indexHistogramBase] < 0){
 			float similarityHistogram = 0;
 			similarityHistogram =
-					(float)HistogramMatchCalculator.getSimilarity(
+					(float)HistogramMatchCalculator.getFractionOverlappingBins(
 							mdhvQueryBlurredHist, indexNode1Query, indexNode2Query, mdhvBaseBlurredHist, indexNode1Base, indexNode2Base);
 			arrSimilarityHistograms[indexHistogramQuery][indexHistogramBase]=similarityHistogram;
 		}