• Home
• com.actelion.research
• openchemlib
• 2024.9.3
• source code
• MoleculeStandardizer.java

×

Project download

Many resources are needed to download a project. Please understand that we have to compensate our server costs. Thank you in advance.
Project price only 1 $

You can buy this project and download/modify it how often you want.

com.actelion.research.chem.MoleculeStandardizer Maven / Gradle / Ivy

package com.actelion.research.chem;

import com.actelion.research.chem.coords.CoordinateInventor;

/**
 * MoleculeStandardizer
 * @author Modest von Korff, Thomas Sander
 * @version 1.0
 * Apr 5, 2012 MvK: Start implementation
 * Oct 2020 MvK,TLS: improved performance and adapted standardization according to following publication:
 * Bento, A. P., Hersey, A., Félix, E., Landrum, G., Gaulton, A., Atkinson, F., ... & Leach, A. R. (2020).
 * An open source chemical structure curation pipeline using RDKit. Journal of Cheminformatics, 12(1), 1-16.

 * Exceptions: - S=O is not transformed into S(+)-O(-)

 *             - If charges (e.g. quaternary nitrogen) cannot be balanced, then Na(+) or Cl(-) are added to neutralize as last resort

 */
public class MoleculeStandardizer {
	public static final int MODE_GET_PARENT = 1;
	public static final int MODE_ADD_NA_AND_CL = 2;

	/**
	 * Under normal circumstances, one should never need to standardize a molecule from an idcode,
	 * because molecules should be standardized before generating the canonical encoding.
	 * An exception is when generating the parent structure using mode MODE_GET_PARENT and
	 * potentially MODE_ADD_NA_AND_CL.
	 * @param idcode
	 * @param coordinates if null the result may change.
	 * @param mode 0 or any combination of MODE_GET_PARENT and MODE_ADD_NA_AND_CL
	 * @return
	 * @throws Exception
	 */
	public static StereoMolecule getStandardized(String idcode, String coordinates, int mode) throws Exception {
		StereoMolecule mol = new IDCodeParser().getCompactMolecule(idcode, coordinates);
		standardize(mol, mode);
		return mol;
	}

	/**
	 * Standardises a molecule and fixes some structural errors.
	 * Typically, this is done before canonicalization.
	 * It includes the following changes:

	 * - different forms of functional groups (e.g. nitro) are normalized to a preferred one

	 * - charged acidic or basic atoms are (de-)protonated to remove charges and neutralize the molecule, if possible.

	 * - alkali/earthalkali/halogene atoms, if charged despite being covalently bound, get uncharged

	 * - trivalent, uncharged oxygens get a positive charge

	 * - unusual amide tautomeric structures, if not in a ring, are inverted

	 * - uncharged isocyano groups get proper charges to validate valences

	 * - wrongly charged azido groups get proper charges to validate valences

	 * - uncharged, quarternary nitrogens get a positive charge

	 * If mode includes MODE_GET_PARENT, then only the largest, normalized fragment is kept.
	 * If mode includes MODE_ADD_NA_AND_CL, then molecules, that are still charged after normalization,
	 * e.g. quarternary ammonium, are neutralized by adding the right amount of Na+ or Cl- ions.
	 * @param mol
	 * @param mode 0 or any combination of MODE_GET_PARENT and MODE_ADD_NA_AND_CL
	 * @throws Exception
	 */
	public static void standardize(StereoMolecule mol, int mode) throws Exception {
		if((mode & MODE_GET_PARENT) != 0) {
			mol.stripSmallFragments();
			mol.stripIsotopInfo();
			}

		repairAndUnify(mol);

		mol.normalizeAmbiguousBonds();

		int remainingCharge = mol.canonizeCharge(true, true);

		if (remainingCharge != 0)
			neutralizeCharges(mol, mode, remainingCharge);

		mol.validateAtomQueryFeatures();
		mol.validateBondQueryFeatures();
	}

	/**
	 * Repairs wrongly uncharged quaternary nitrogen. Unifies carbonyl acid groups,
	 * sulfonic acid, phosphoric acid, phenolic oxygen. Means: negative charges are removed.
	 * Adds Na+ or Cl- for final charge equilibration.
	 * @param mol
	 */
	private static void repairAndUnify(StereoMolecule mol) {
		mol.ensureHelperArrays(Molecule.cHelperRings);

		repairCovalentBoundChargedAlkaliAndHalogen(mol);
		chargeTrivalentOxygen(mol);
		repairBadAmideTautomer(mol);
		repairQuaternaryNitrogen(mol);
		unifyIsoCyano(mol);
		unifyAzido(mol);
	}

	private static void neutralizeCharges(StereoMolecule mol, int mode, int totalCharge) {
		mol.ensureHelperArrays(Molecule.cHelperNeighbours);

		for (int atom=0; atom0; atom++) {
			if (AtomFunctionAnalyzer.isAcidicOxygen(mol, atom)) {
				mol.setAtomCharge(atom, -1);
				totalCharge--;
			}
		}

		for (int atom=0; atom 0) {
				int ind = mol.addAtom(17);
				mol.setAtomCharge(ind, -1);
				totalCharge--;
			}

			// add Na+
			while (totalCharge < 0) {
				int ind = mol.addAtom(11);
				mol.setAtomCharge(ind, 1);
				totalCharge++;
			}

			new CoordinateInventor(CoordinateInventor.MODE_KEEP_MARKED_ATOM_COORDS
					+ CoordinateInventor.MODE_REMOVE_HYDROGEN).invent(mol);
		}
	}

	/**
	 * Remove wrong charges on halogen and (earth)alkali atoms, if they are
	 * covalently bound.
	 * @param mol
	 */
	private static void repairCovalentBoundChargedAlkaliAndHalogen(StereoMolecule mol) {
		for (int atom=0; atom