/// <summary>
/// RemovePseudoAtoms
/// </summary>
/// <param name="mol"></param>
/// <returns>Pseudo atom count</returns>
public static int RemovePseudoAtoms(IAtomContainer mol)
{
List <IAtom> pseudoAtoms = new List <IAtom>();
for (int ai = 0; ai < mol.Atoms.Count; ai++)
{
IAtom atom = mol.Atoms[ai];
if (atom is IPseudoAtom)
{
pseudoAtoms.Add(atom);
}
}
if (pseudoAtoms.Count == 0)
{
return(0);
}
foreach (IAtom atom in pseudoAtoms)
{
mol.RemoveAtom(atom);
}
AtomContainerManipulator.PercieveAtomTypesAndConfigureAtoms(mol);
mol = AtomContainerManipulator.SuppressHydrogens(mol);
GetHydrogenAdder().AddImplicitHydrogens(mol);
return(pseudoAtoms.Count);
}
public override void TestBug706786()
{
IAtomContainer superStructure = Bug706786_1();
IAtomContainer subStructure = Bug706786_2();
AddImplicitHydrogens(superStructure);
AddImplicitHydrogens(subStructure);
// SMARTS is now correct and D will include H atoms, CDK had this wrong
// for years (had it has non-H count). Whilst you can set the optional
// SMARTS flavor CDK_LEGACY this is not correct
AtomContainerManipulator.SuppressHydrogens(superStructure);
AtomContainerManipulator.SuppressHydrogens(subStructure);
IFingerprinter fpr = GetBitFingerprinter();
IBitFingerprint superBits = fpr.GetBitFingerprint(superStructure);
IBitFingerprint subBits = fpr.GetBitFingerprint(subStructure);
Assert.IsTrue(BitArrays.Equals(
AsBitSet(0, 11, 13, 17, 40, 48, 136, 273, 274, 278, 286, 294, 299, 301, 304, 306),
superBits.AsBitSet()));
Assert.IsTrue(BitArrays.Equals(
AsBitSet(1, 17, 273, 274, 278, 294, 306),
subBits.AsBitSet()));
}
/// <summary>
/// Remove the following features from a molecule
/// - Atom non-standard mass
/// - Stereo chemistry
/// - explicit hydrogens
/// </summary>
/// <param name="src"></param>
/// <returns>Modified mol</returns>
public static IAtomContainer RemoveIsotopesStereoExplicitHydrogens(
IAtomContainer src)
{
IAtom[] atoms = new IAtom[src.Atoms.Count];
IBond[] bonds = new IBond[src.Bonds.Count];
IChemObjectBuilder builder = src.Builder;
for (int i = 0; i < atoms.Length; i++)
{
IAtom atom = src.Atoms[i];
IAtom atom2 = builder.NewAtom(atom.Symbol);
SetImplicitHydrogenCount(atom2, GetImplicitHydrogenCount(atom));
atom2.Point2D = atom.Point2D;
atoms[i] = atom2;
}
for (int i = 0; i < bonds.Length; i++)
{
IBond bond = src.Bonds[i];
int u = bond.Atoms[0].Index;
int v = bond.Atoms[1].Index;
IBond bond2 = builder.NewBond(atoms[u], atoms[v]);
bond2.IsAromatic = bond.IsAromatic;
bond2.IsInRing = bond.IsInRing;
bond2.Order = bond.Order;
bond2.IsAromatic = bond.IsAromatic;
bond2.IsSingleOrDouble = bond.IsSingleOrDouble;
bonds[i] = bond2;
}
IAtomContainer dest = ChemObjectBuilder.Instance.NewAtomContainer();
dest.SetAtoms(atoms);
dest.SetBonds(bonds);
AtomContainerManipulator.PercieveAtomTypesAndConfigureAtoms(dest);
dest = AtomContainerManipulator.SuppressHydrogens(dest);
GetHydrogenAdder().AddImplicitHydrogens(dest);
return(dest);
}
/// <summary>
/// Overloaded <see cref="GetTautomers(IAtomContainer)"/> to get tautomers for an input molecule with the InChI already
/// provided as input argument.
/// </summary>
/// <param name="mol">and input molecule for which to generate tautomers</param>
/// <param name="inchi">InChI for the input molecule</param>
/// <param name="amap">ordering of the molecules atoms in the InChI</param>
/// <returns>a list of tautomers</returns>
/// <exception cref="CDKException"></exception>
private List <IAtomContainer> GetTautomers(IAtomContainer mol, string inchi, long[] amap)
{
//Initial checks
if (mol == null || inchi == null)
{
throw new CDKException("Please provide a valid input molecule and its corresponding InChI value.");
}
// shallow copy since we will suppress hydrogens
mol = mol.Builder.NewAtomContainer(mol);
var tautomers = new List <IAtomContainer>();
if (!inchi.Contains("(H"))
{ //No mobile H atoms according to InChI, so bail out.
tautomers.Add(mol);
return(tautomers);
}
//Preparation: translate the InChi
var inchiAtomsByPosition = GetElementsByPosition(inchi, mol);
var inchiMolGraph = ConnectAtoms(inchi, mol, inchiAtomsByPosition);
if (amap != null && amap.Length == mol.Atoms.Count)
{
for (int i = 0; i < amap.Length; i++)
{
mol.Atoms[i].Id = amap[i].ToString(NumberFormatInfo.InvariantInfo);
}
mol = AtomContainerManipulator.SuppressHydrogens(mol);
}
else
{
mol = AtomContainerManipulator.SuppressHydrogens(mol);
MapInputMoleculeToInChIMolgraph(inchiMolGraph, mol);
}
var mobHydrAttachPositions = new List <int>();
var totalMobHydrCount = ParseMobileHydrogens(mobHydrAttachPositions, inchi);
tautomers = ConstructTautomers(mol, mobHydrAttachPositions, totalMobHydrCount);
//Remove duplicates
return(RemoveDuplicates(tautomers));
}
public override void TestBug706786()
{
IAtomContainer superStructure = Bug706786_1();
IAtomContainer subStructure = Bug706786_2();
AtomContainerManipulator.PercieveAtomTypesAndConfigureAtoms(superStructure);
AtomContainerManipulator.PercieveAtomTypesAndConfigureAtoms(subStructure);
AddImplicitHydrogens(superStructure);
AddImplicitHydrogens(subStructure);
// SMARTS is now correct and D will include H atoms, CDK had this wrong
// for years (had it has non-H count). Whilst you can set the optional
// SMARTS flavor CDK_LEGACY this is not correct
AtomContainerManipulator.SuppressHydrogens(superStructure);
AtomContainerManipulator.SuppressHydrogens(subStructure);
IFingerprinter fpr = new EStateFingerprinter();
IBitFingerprint superBits = fpr.GetBitFingerprint(superStructure);
IBitFingerprint subBits = fpr.GetBitFingerprint(subStructure);
Assert.IsTrue(BitArrays.Equals(AsBitSet(6, 11, 12, 15, 16, 18, 33, 34, 35), superBits.AsBitSet()));
Assert.IsTrue(BitArrays.Equals(AsBitSet(8, 11, 16, 35), subBits.AsBitSet()));
}
/// <summary>
/// Integrate MMP difference and common fragments
/// </summary>
/// <param name="smilesFrags"></param>
/// <returns>Molfile of integrated mol with difference fragment hilighted via special atom isotope values </returns>
public string IntegrateAndHilightMmpStructure(
string smilesFrags)
{
IAtomContainer mol;
IAtom atom;
string frag, context1, context2, molfile, chime = "", molfile2 = "";
int ai, atomCount;
//IsotopeConversionTest(""); // debug
DateTime t0 = DateTime.Now;
try
{
Lex.Split(smilesFrags, ":", out frag, out context1, out context2);
if (Lex.Eq(context2, "NULL"))
{
context2 = ""; // fixup for undefined Redshift fragment smiles fragment represented as "NULL" rather than a blank string
}
mol = SmilesToAtomContainer(frag);
atomCount = mol.Atoms.Count;
for (ai = 0; ai < atomCount; ai++) // set isotope labels to mark difference frag
{
atom = mol.Atoms[ai];
int mn = GetMassNumber(atom);
if (mn == 0)
{
SetMassNumber(atom, hilightIsotopeValue); // mark with special value
}
else if (mn > 0)
{
SetMassNumber(atom, -mn); // negate original value to indicate part of difference frag
}
}
if (Lex.IsDefined(context1))
{
JoinMmpFragments(mol, context1, 1);
atomCount = mol.Atoms.Count;
}
if (Lex.IsDefined(context2))
{
JoinMmpFragments(mol, context2, 2);
atomCount = mol.Atoms.Count;
}
mol = AtomContainerManipulator.SuppressHydrogens(mol); // fix potential valency issues
atomCount = mol.Atoms.Count;
for (ai = 0; ai < atomCount; ai++) // clean up structure for good display
{
atom = mol.Atoms[ai];
if (atom.Valency != null && atom.Valency.Value != 0)
{
atom.Valency = 0;
}
if (GetImplicitHydrogenCount(atom) != 0)
{
SetImplicitHydrogenCount(atom, 0);
}
}
AtomContainerManipulator.PercieveAtomTypesAndConfigureAtoms(mol);
mol = AtomContainerManipulator.SuppressHydrogens(mol);
GetHydrogenAdder().AddImplicitHydrogens(mol);
mol = GenerateCoordinates(mol);
atomCount = mol.Atoms.Count;
molfile = ConvertIsotopeValuesToHilighting(mol); // about 3.0 ms per IntegrateAndHilightMmpStructure call
double msTime = TimeOfDay.Delta(t0);
IntegrateAndHilightMmpStructureCount++;
IntegrateAndHilightMmpStructureTotalTime += msTime;
IntegrateAndHilightMmpStructureAvgTime = IntegrateAndHilightMmpStructureTotalTime / IntegrateAndHilightMmpStructureCount;
return(molfile);
}
catch (Exception ex)
{
string msg = "Error integrating Smiles Fragments: " + smilesFrags + DebugLog.FormatExceptionMessage(ex, true);
DebugLog.Message(msg);
return("");
}
}