Do not leave hydrogen counts unset - also load isotope information.

Signed-off-by: Egon Willighagen <egonw@users.sourceforge.net>
cdk · Dec 10, 2013 · 5ad5e0e · 5ad5e0e
1 parent bf9b180
commit 5ad5e0e
Show file tree

Hide file tree

Showing 2 changed files with 56 additions and 14 deletions.
diff --git a/src/main/org/openscience/cdk/inchi/InChIToStructure.java b/src/main/org/openscience/cdk/inchi/InChIToStructure.java
@@ -20,6 +20,7 @@
  */
 package org.openscience.cdk.inchi;
 
+import java.io.IOException;
 import java.util.HashMap;
 import java.util.List;
 import java.util.Map;
@@ -40,6 +41,7 @@
 import org.openscience.cdk.CDKConstants;
 import org.openscience.cdk.annotations.TestClass;
 import org.openscience.cdk.annotations.TestMethod;
+import org.openscience.cdk.config.Isotopes;
 import org.openscience.cdk.exception.CDKException;
 import org.openscience.cdk.interfaces.IAtom;
 import org.openscience.cdk.interfaces.IAtomContainer;
@@ -92,6 +94,9 @@ public class InChIToStructure {
 
     protected IAtomContainer molecule;
 
+    // magic number - indicates isotope mass is relative
+    private static final int ISOTOPIC_SHIFT_FLAG = 10000;
+
     /**
      * Constructor. Generates CDK AtomContainer from InChI.
      * @param inchi
@@ -163,23 +168,34 @@ protected void generateAtomContainerFromInchi(IChemObjectBuilder builder) throws
             inchiCdkAtomMap.put(iAt, cAt);
 
             cAt.setID("a" + i);
-            cAt.setSymbol(iAt.getElementType());
+            cAt.setSymbol(iAt.getElementType());                
             cAt.setAtomicNumber(PeriodicTable.getAtomicNumber(cAt.getSymbol()));
 
-            // Ignore coordinates - all zero
-
-            int charge = iAt.getCharge();
-            if (charge != 0) {
-                cAt.setFormalCharge(charge);
-            }
+            // Ignore coordinates - all zero - unless aux info was given... but
+            // the CDK doesn't have an API to provide that
+
+            // InChI does not have unset properties so we set charge, 
+            // hydrogen count (implicit) and isotopic mass
+            cAt.setFormalCharge(iAt.getCharge());
+            cAt.setImplicitHydrogenCount(iAt.getImplicitH());
 
-            // hydrogenCount contains number of implict hydrogens, not
-            // total number
-            // Ref: Posting to cdk-devel list by Egon Willighagen 2005-09-17
-            int numH = iAt.getImplicitH();
-            if (numH != 0) {
-                cAt.setImplicitHydrogenCount(numH);
-            }
+            int isotopicMass = iAt.getIsotopicMass();
+
+            if (isotopicMass != 0) {
+                if (ISOTOPIC_SHIFT_FLAG == (isotopicMass & ISOTOPIC_SHIFT_FLAG)) {
+                    try {
+                        int massNumber = Isotopes.getInstance()
+                                           .getMajorIsotope(cAt.getAtomicNumber())
+                                           .getMassNumber();
+                        cAt.setMassNumber(massNumber + (isotopicMass - ISOTOPIC_SHIFT_FLAG));
+                    } catch (IOException e) {
+                        throw new CDKException("Could not load Isotopes data", e);
+                    }
+                }
+                else {
+                    cAt.setMassNumber(isotopicMass);     
+                }
+            } 
 
             molecule.addAtom(cAt);
         }

diff --git a/src/test/org/openscience/cdk/inchi/InChIToStructureTest.java b/src/test/org/openscience/cdk/inchi/InChIToStructureTest.java
@@ -31,6 +31,9 @@
 import org.openscience.cdk.silent.AtomContainer;
 import org.openscience.cdk.silent.SilentChemObjectBuilder;
 
+import static org.hamcrest.CoreMatchers.instanceOf;
+import static org.hamcrest.CoreMatchers.is;
+import static org.hamcrest.CoreMatchers.notNullValue;
 import static org.junit.Assert.assertNotNull;
 
 /**
@@ -158,5 +161,28 @@ public void testGetAtomContainer_IChemObjectBuilder() throws CDKException {
 		// OK, this is not typical use, but maybe the above generate method should be private
 		Assert.assertTrue(container instanceof AtomContainer);
 	}
+
+    @Test public void atomicOxygen() throws CDKException {
+        InChIToStructure parser = new InChIToStructure(
+                "InChI=1S/O", DefaultChemObjectBuilder.getInstance()
+        );
+        parser.generateAtomContainerFromInchi(SilentChemObjectBuilder.getInstance());
+        IAtomContainer container = parser.getAtomContainer();
+        Assert.assertThat(container, is(instanceOf(AtomContainer.class)));
+        Assert.assertThat(container.getAtom(0).getImplicitHydrogenCount(), is(notNullValue()));
+        Assert.assertThat(container.getAtom(0).getImplicitHydrogenCount(), is(0));
+    }  
+
+    @Test public void heavyOxygenWater() throws CDKException {
+        InChIToStructure parser = new InChIToStructure(
+                "InChI=1S/H2O/h1H2/i1+2", DefaultChemObjectBuilder.getInstance()
+        );
+        parser.generateAtomContainerFromInchi(SilentChemObjectBuilder.getInstance());
+        IAtomContainer container = parser.getAtomContainer();
+        Assert.assertThat(container, is(instanceOf(AtomContainer.class)));
+        Assert.assertThat(container.getAtom(0).getImplicitHydrogenCount(), is(notNullValue()));
+        Assert.assertThat(container.getAtom(0).getImplicitHydrogenCount(), is(2));
+        Assert.assertThat(container.getAtom(0).getMassNumber(), is(18));
+    }
 
 }