Ejemplo n.º 1
0
    def buildOeBinaryMolCacheFromIndex(self,
                                       filePath,
                                       ccIdxD,
                                       quietFlag=False,
                                       fpTypeList=None,
                                       limitPerceptions=False,
                                       suppressHydrogens=False):
        """Build cache of OEGraphMol() objects from the input chemical component search index.

        Args:
            filePath (str): output cache file path
            ccIdxD (dict): search index dictionary
            quietFlag (bool, optional): suppress OE output. Defaults to False.
            fpTypeList (list, optional): list of fingerprint types. Defaults to None.
            limitPerceptions (bool, optional): suppress automatic chemical perceptions. Defaults to False.
            suppressHydrogens (bool, optional): suppress explicit hydrogen count. Defaults to False.

        Returns:
            (int, int, list): chem comp success count, error count, chem comp identifier failure list
        """
        failIdList = []
        ccCount = 0
        errCount = 0
        startTime = time.time()
        try:
            ofs = oechem.oemolostream()
            ofs.SetFormat(oechem.OEFormat_OEB)
            if ofs.open(filePath):
                oemf = OeMoleculeFactory()
                if quietFlag:
                    oemf.setQuiet()
                for searchCcId, ccIdx in ccIdxD.items():
                    oemf.setDescriptor(ccIdx["smiles"], "oe-iso-smiles",
                                       searchCcId)
                    ok = oemf.build(molBuildType="oe-iso-smiles",
                                    limitPerceptions=limitPerceptions)
                    if ok and fpTypeList:
                        fpOk = oemf.addFingerPrints(fpTypeList)
                        if not fpOk:
                            logger.info("Fingerprint generation fails for %r",
                                        searchCcId)
                    if ok:
                        if not suppressHydrogens:
                            oemf.addExplicitHydrogens()
                            oemf.setSimpleAtomNames()
                        oeMol = oemf.getMol(
                            suppressHydrogens=suppressHydrogens)
                        oechem.OEWriteMolecule(ofs, oeMol)
                        ccCount += 1
                    if not ok:
                        # build failed incomplete component (e.g. missing atoms or bonds)
                        errCount += 1
                        failIdList.append(searchCcId)
            else:
                logger.error("Unable to open cache database %s", filePath)
                errCount += 1
        except Exception as e:
            logger.exception("Failing with %s", str(e))
        #
        endTime = time.time()
        logger.info("Completed operation at %s (%.4f seconds)",
                    time.strftime("%Y %m %d %H:%M:%S", time.localtime()),
                    endTime - startTime)
        return ccCount, errCount, failIdList
Ejemplo n.º 2
0
    def buildOeBinaryMolCache(self,
                              filePath,
                              ccObjD,
                              molBuildType="model-xyz",
                              quietFlag=False,
                              fpTypeList=None,
                              limitPerceptions=False,
                              suppressHydrogens=False):
        """Build cache of OEMol() objects from the input chemical component definition list.

        Args:
            filePath (str): output cache file path
            ccObjD (dict):  chemical component object dictionary
            molBuildType (str, optional): [description]. Defaults to "model-xyz".
            quietFlag (bool, optional): [description]. Defaults to False.
            fpTypeList (list, optional): fingerprint type list. Defaults to None.
            limitPerceptions (bool, optional): suppress automatic chemical perceptions. Defaults to False.
            suppressHydrogens (bool, optional): suppress explicit hydrogen count. Defaults to False.

        Returns:
            (int, int, list): chem comp success count, error count, chem comp identifier failure list

        """
        ok = False
        startTime = time.time()
        failIdList = []
        ccCount = 0
        errCount = 0
        try:
            ofs = oechem.oemolostream()
            ofs.SetFormat(oechem.OEFormat_OEB)
            if ofs.open(filePath):
                oemf = OeMoleculeFactory()
                if quietFlag:
                    oemf.setQuiet()
                for ccId, ccObj in ccObjD.items():
                    tId = oemf.setChemCompDef(ccObj)
                    if tId and tId == ccId:
                        ok = oemf.build(molBuildType=molBuildType,
                                        limitPerceptions=limitPerceptions)
                        if ok and fpTypeList:
                            fpOk = oemf.addFingerPrints(fpTypeList)
                            if not fpOk:
                                logger.info(
                                    "Fingerprint generation fails for %r",
                                    ccId)
                        if ok:
                            oeMol = oemf.getMol(
                                suppressHydrogens=suppressHydrogens)
                            oechem.OEWriteMolecule(ofs, oeMol)
                            ccCount += 1
                    if not ok or not tId:
                        # build failed incomplete component (e.g. missing atoms or bonds)
                        errCount += 1
                        failIdList.append(ccId)
            else:
                logger.error("Unable to open cache database %s", filePath)
                errCount += 1
        except Exception as e:
            logger.exception("Failing with %s", str(e))
        #
        endTime = time.time()
        logger.info("Completed operation at %s (%.4f seconds)",
                    time.strftime("%Y %m %d %H:%M:%S", time.localtime()),
                    endTime - startTime)
        return ccCount, errCount, failIdList