void testSpaceGroupClean()
{
// See https://github.com/openbabel/openbabel/pull/254
OBConversion conv;
OBMol mol;
conv.SetInFormat("cif");
conv.SetOutFormat("pdb");
conv.ReadFile(&mol, GetFilename("test02.cif"));
OBUnitCell* pUC = (OBUnitCell*)mol.GetData(OBGenericDataType::UnitCell);
const SpaceGroup* pSG = pUC->GetSpaceGroup();
SpaceGroup* sg = new SpaceGroup(*pSG);
pSG = SpaceGroup::Find(sg);
OB_ASSERT( pSG != NULL );
// Check also for errors and warnings
string summary = obErrorLog.GetMessageSummary();
OB_ASSERT( summary.find("error") == string::npos);
OB_ASSERT( summary.find("warning") == string::npos);
OB_ASSERT( pSG->GetId() == 166 );
string pdb = conv.WriteString(&mol);
pdb = conv.WriteString(&mol);
OB_ASSERT(pdb.find("H -3 m") != string::npos);
}
void testPdbOccupancies()
{
// See https://github.com/openbabel/openbabel/pull/1558
OBConversion conv;
OBMol mol;
conv.SetInFormat("cif");
conv.SetOutFormat("pdb");
conv.ReadFile(&mol, GetFilename("test08.cif"));
string pdb = conv.WriteString(&mol);
conv.AddOption("o", OBConversion::OUTOPTIONS);
pdb = conv.WriteString(&mol);
OB_ASSERT(pdb.find("HETATM 1 NA UNL 1 0.325 0.000 4.425 0.36") != string::npos);
OB_ASSERT(pdb.find("HETATM 17 O UNL 8 1.954 8.956 3.035 1.00") != string::npos);
OBMol mol_pdb;
conv.SetInFormat("pdb");
conv.ReadFile(&mol_pdb, GetFilename("test09.pdb"));
pdb = conv.WriteString(&mol_pdb);
OB_ASSERT(pdb.find("HETATM 1 NA UNL 1 0.325 0.000 4.425 0.36") != string::npos);
OB_ASSERT(pdb.find("HETATM 2 NA UNL 1 0.002 8.956 1.393 0.10") != string::npos);
OB_ASSERT(pdb.find("HETATM 17 O UNL 8 1.954 8.956 3.035 1.00") != string::npos);
}
// A segfault was occuring when a Universal SMILES was output after an InChIfied SMILES.
// This was due to short-circuit caching of InChIs on reading. The fix was to limit
// the situations when the cached value was used, but also to delete the cached value
// in this particular instance.
void test_Issue135_UniversalSmiles()
{
// Test writing U smiles after I smiles
OBConversion conv;
conv.SetInFormat("smi");
OBMol mol;
conv.ReadString(&mol, "C(=O)([O-])C(=O)O");
conv.SetOutFormat("smi");
conv.SetOptions("I", OBConversion::OUTOPTIONS);
std::string res = conv.WriteString(&mol, true);
OB_COMPARE(res, "C(=O)(C(=O)O)[O-]");
conv.SetOptions("U", OBConversion::OUTOPTIONS);
res = conv.WriteString(&mol, true);
OB_COMPARE(res, "C(=O)(C(=O)[O-])O");
}
void testPdbRemSpacesHMName()
{
// See https://github.com/openbabel/openbabel/pull/1558
OBConversion conv;
OBMol mol;
conv.SetInFormat("cif");
conv.SetOutFormat("pdb");
conv.ReadFile(&mol, GetFilename("test07.cif"));
string pdb = conv.WriteString(&mol);
conv.AddOption("o", OBConversion::OUTOPTIONS);
pdb = conv.WriteString(&mol);
OB_ASSERT(pdb.find("I41/amd:2") != string::npos);
}
void testPdbOutHexagonalAlternativeOrigin2()
{
// See https://github.com/openbabel/openbabel/pull/1558
OBConversion conv;
OBMol mol;
conv.SetInFormat("cif");
conv.SetOutFormat("pdb");
conv.ReadFile(&mol, GetFilename("test06.cif"));
string pdb = conv.WriteString(&mol);
conv.AddOption("o", OBConversion::OUTOPTIONS);
pdb = conv.WriteString(&mol);
OB_ASSERT(pdb.find("H -3 m") != string::npos);
}
int main() {
OBAtom a, b, c;
a.SetAtomicNum(8);
b.SetAtomicNum(6);
c.SetAtomicNum(8);
OBMol mol;
mol.AddAtom(a);
mol.AddAtom(b);
mol.AddAtom(c);
mol.AddBond(1,2,2);
mol.AddBond(2,3,2);
OBConversion conv;
conv.SetOutFormat("SMI");
cout << conv.WriteString(&mol,1) << endl;
OBSmartsPattern sp;
sp.Init ("C~*");
sp.Match (mol,false);
cout << sp.NumMatches() << endl;
cout << sp.GetUMapList().size() << endl;
return EXIT_SUCCESS;
}
void testPdbOutAlternativeOrigin()
{
// See https://github.com/openbabel/openbabel/pull/1558
OBConversion conv;
OBMol mol;
conv.SetInFormat("cif");
conv.SetOutFormat("pdb");
conv.ReadFile(&mol, GetFilename("test04.cif"));
string pdb = conv.WriteString(&mol);
// ending space is needed to check that there is no origin set
OB_ASSERT(pdb.find("P 4/n b m ") != string::npos);
conv.AddOption("o", OBConversion::OUTOPTIONS);
pdb = conv.WriteString(&mol);
OB_ASSERT(pdb.find("P 4/n b m:1") != string::npos);
}
// Reading an InChI and then adding hydrogens messed up the structure
void test_Issue134_InChI_addH()
{
OBConversion conv;
conv.SetInFormat("inchi");
OBMol mol;
conv.ReadString(&mol, "InChI=1S/C2H7NO/c1-2(3)4/h2,4H,3H2,1H3/t2-/m0/s1");
OB_ASSERT(!mol.HasData(OBGenericDataType::VirtualBondData));
mol.AddHydrogens();
conv.SetOutFormat("smi");
std::string res = conv.WriteString(&mol, true);
OB_COMPARE(res, "C[C@@H](N)O");
}
void testCIFMolecules()
{
// See https://github.com/openbabel/openbabel/pull/1558
OBConversion conv;
OBMol mol;
conv.SetInFormat("cif");
conv.SetOutFormat("smi"); // check for disconnected fragments
conv.ReadFile(&mol, GetFilename("1519159.cif"));
string smi = conv.WriteString(&mol);
// never, never disconnected fragments from a molecule
OB_ASSERT(smi.find(".") == string::npos);
}
int main(int argc, char **argv)
{
// Define location of file formats for testing
#ifdef FORMATDIR
char env[BUFF_SIZE];
snprintf(env, BUFF_SIZE, "BABEL_LIBDIR=%s", FORMATDIR);
putenv(env);
#endif
std::ifstream ifs(GetFilename("canonstable.can").c_str());
OB_REQUIRE( ifs );
OBMol mol;
OBConversion conv;
conv.SetInFormat("smi");
conv.SetOutFormat("can");
std::string line;
while (std::getline(ifs, line)) {
OB_REQUIRE( conv.ReadString(&mol, line.c_str()) );
std::vector<OBAtom*> atoms;
FOR_ATOMS_OF_MOL(atom, mol)
atoms.push_back(&*atom);
for (int i = 0; i < 5; ++i) {
// shuffle the atoms
std::random_shuffle(atoms.begin(), atoms.end());
mol.RenumberAtoms(atoms);
// get can smiles
mol.SetTitle("");
std::string cansmi = conv.WriteString(&mol, true);
// comapare with ref
if (cansmi != line) {
cout << "ref = " << line << endl;
cout << "can = " << cansmi << endl;
OB_ASSERT( cansmi == line );
}
}
}
return 0;
}
bool OpReadConformers::ProcessVec(std::vector<OBBase*>& vec)
{
// DeferredFormat collects all the molecules, they are processed here, and Deferred Format outputs them
OBConversion smconv;
smconv.AddOption("n");
if(!smconv.SetOutFormat("smi"))
{
obErrorLog.ThrowError(__FUNCTION__, "SmilesFormat is not loaded" , obError, onceOnly);
return false;
}
std::string smiles, stored_smiles;
OBMol* stored_pmol=NULL;
std::vector<OBBase*>::iterator iter;
for(iter= vec.begin();iter!=vec.end();++iter)
{
OBMol* pmol = dynamic_cast<OBMol*>(*iter);
if(!pmol)
continue;
smiles = smconv.WriteString(pmol);
Trim(smiles);
if(stored_smiles==smiles)
{
//add the coordinates of the current mol to the stored one as a conformer, and delete current mol
double *confCoord = new double [pmol->NumAtoms() * 3];
memcpy((char*)confCoord,(char*)pmol->GetCoordinates(),sizeof(double)*3*pmol->NumAtoms());
stored_pmol->AddConformer(confCoord);
delete pmol;
*iter = NULL;
}
else
{
stored_pmol = pmol;
stored_smiles = smiles;
}
}
//erase the NULLS
vec.erase(std::remove(vec.begin(),vec.end(), (void*)NULL), vec.end());
return true;
}
bool FastSearchFormat::ObtainTarget(OBConversion* pConv, vector<OBMol>& patternMols, const string& indexname)
{
//Obtains an OBMol from:
// the filename in the -s option or
// the SMARTS string in the -s option or
// by converting the file in the -S or -aS options (deprecated).
// If there is no -s -S or -aS option, information on the index file is displayed.
OBMol patternMol;
patternMol.SetIsPatternStructure();
const char* p = pConv->IsOption("s",OBConversion::GENOPTIONS);
bool OldSOption=false;
//If no -s option, make OBMol from file in -S option or -aS option (both deprecated)
if(!p)
{
p = pConv->IsOption("S",OBConversion::GENOPTIONS);
if(!p)
p = pConv->IsOption("S",OBConversion::INOPTIONS);//for GUI mainly
OldSOption = true;
}
if(p)
{
vector<string> vec;
tokenize(vec, p);
//ignore leading ~ (not relevant to fastsearch)
if(vec[0][0]=='~')
vec[0].erase(0,1);
if(vec.size()>1 && vec[1]=="exact")
pConv->AddOption("e", OBConversion::INOPTIONS);
OBConversion patternConv;
OBFormat* pFormat;
//Interpret as a filename if possible
string& txt =vec [0];
if( txt.empty() ||
txt.find('.')==string::npos ||
!(pFormat = patternConv.FormatFromExt(txt.c_str())) ||
!patternConv.SetInFormat(pFormat) ||
!patternConv.ReadFile(&patternMol, txt) ||
patternMol.NumAtoms()==0)
//if false, have a valid patternMol from a file
{
//is SMARTS/SMILES
//Replace e.g. [#6] in SMARTS by C so that it can be converted as SMILES
//for the fingerprint phase, but allow more generality in the SMARTS phase.
for(;;)
{
string::size_type pos1, pos2;
pos1 = txt.find("[#");
if(pos1==string::npos)
break;
pos2 = txt.find(']');
int atno;
if(pos2!=string::npos && (atno = atoi(txt.substr(pos1+2, pos2-pos1-2).c_str())) && atno>0)
txt.replace(pos1, pos2-pos1+1, etab.GetSymbol(atno));
else
{
obErrorLog.ThrowError(__FUNCTION__,"Ill-formed [#n] atom in SMARTS", obError);
return false;
}
}
bool hasTildeBond;
if( (hasTildeBond = (txt.find('~')!=string::npos)) ) // extra parens to indicate truth value
{
//Find ~ bonds and make versions of query molecule with a single and aromatic bonds
//To avoid having to parse the SMILES here, replace ~ by $ (quadruple bond)
//and then replace this in patternMol. Check first that there are no $ already
//Sadly, isocynanides may have $ bonds.
if(txt.find('$')!=string::npos)
{
obErrorLog.ThrowError(__FUNCTION__,
"Cannot use ~ bonds in patterns with $ (quadruple) bonds.)", obError);
return false;
}
replace(txt.begin(),txt.end(), '~' , '$');
}
//read as standard SMILES
patternConv.SetInFormat("smi");
if(!patternConv.ReadString(&patternMol, vec[0]))
{
obErrorLog.ThrowError(__FUNCTION__,"Cannot read the SMILES string",obError);
return false;
}
if(hasTildeBond)
{
AddPattern(patternMols, patternMol, 0); //recursively add all combinations of tilde bond values
return true;
}
}
else
{
// target(s) are in a file
patternMols.push_back(patternMol);
while(patternConv.Read(&patternMol))
patternMols.push_back(patternMol);
return true;
}
}
if(OldSOption) //only when using deprecated -S and -aS options
{
//make -s option for later SMARTS test
OBConversion conv;
if(conv.SetOutFormat("smi"))
{
string optiontext = conv.WriteString(&patternMol, true);
pConv->AddOption("s", OBConversion::GENOPTIONS, optiontext.c_str());
}
}
if(!p)
{
//neither -s or -S options provided. Output info rather than doing search
const FptIndexHeader& header = fs.GetIndexHeader();
string id(header.fpid);
if(id.empty())
id = "default";
clog << indexname << " is an index of\n " << header.datafilename
<< ".\n It contains " << header.nEntries
<< " molecules. The fingerprint type is " << id << " with "
<< OBFingerprint::Getbitsperint() * header.words << " bits.\n"
<< "Typical usage for a substructure search:\n"
<< "obabel indexfile.fs -osmi -sSMILES\n"
<< "(-s option in GUI is 'Convert only if match SMARTS or mols in file')" << endl;
return false;
}
patternMols.push_back(patternMol);
return true;
}
int main(int argc,char *argv[])
{
// turn off slow sync with C-style output (we don't use it anyway).
std::ios::sync_with_stdio(false);
if (argc != 1)
{
cout << "Usage: conversion" << endl;
cout << " Unit tests for OBConversion " << endl;
return(-1);
}
cout << "# Unit tests for OBConversion \n";
// the number of tests for "prove"
cout << "1..9\n";
cout << "ok 1\n"; // for loading tests
OBMol obMol;
OBConversion obConversion;
obConversion.SetInAndOutFormats("smi", "mdl");
cout << "ok 2\n";
obConversion.ReadString(&obMol, "C1=CC=CS1");
cout << "ok 3\n";
if (obMol.NumAtoms() == 5) {
cout << "ok 4\n";
} else {
cout << "not ok 4\n";
}
obMol.AddHydrogens();
if (obMol.NumAtoms() == 9) {
cout << "ok 5\n";
} else {
cout << "not ok 5\n";
}
if ( (obConversion.WriteString(&obMol)).length() > 0)
cout << "ok 6\n";
else
cout << "not ok 6\n";
// PR#1474265
obConversion.WriteFile(&obMol, "test.mdl");
ifstream ifs("test.mdl");
if (ifs.good())
cout << "ok 7\n";
else
cout << "not ok 7\n";
// PR#143577
obConversion.SetInFormat("mdl");
obConversion.ReadFile(&obMol, "test.mdl");
if ( remove("test.mdl") != -1)
cout << "ok 8\n";
else
cout << "not ok 8\n";
// gzip input
// gzip output
// multi-molecule reading
// PR#1465586
// aromatics.smi
// attype.00.smi
//ReadFile()
//Read()
//WriteString()
// GetOutputIndex()
// IsLast
//ReadString()
//IsFirstInput
//Read()
// splitting
// splitting using gzip-input
// PR#1357705
// size 0 input
// PR#1250900
// RegisterFormat
// FindFormat
// FormatFromExt
// FormatFromMIME
// GetNextFormat
// GetDefaultFormat
// BatchFileName
// IncrementedFileName
// option handling
// AddOption
// IsOption
// RemoveOption
// IsOption
// SetOptions
// IsOption
// RegisterOptionParam
// GetOptionParams
// GetInStream
// GetOutStream
// SetInStream
// SetOutStream
// nasty tests
obConversion.ReadString(&obMol, "");
obConversion.Read(&obMol);
cout << "ok 9\n";
return(0);
}
bool OpNewS::Do(OBBase* pOb, const char* OptionText, OpMap* pmap, OBConversion* pConv)
{
OBMol* pmol = dynamic_cast<OBMol*>(pOb);
if(!pmol)
return false;
// The SMARTS and any other parameters are extracted on the first molecule
// and stored in the member variables. The parameter is cleared so that
// the original -s option in transform.cpp is inactive
//string txt(pmap->find(GetID())->second); // ID can be "s" or "v"
vector<OBQuery*>::iterator qiter;
if(OptionText && *OptionText)//(!pConv || pConv->IsFirstInput())
{
//Set up on first call
queries.clear();
query=NULL;
nPatternAtoms=0;
inv=false;
tokenize(vec, OptionText);
inv = GetID()[0]=='v';
if(vec[0][0]=='~')
{
inv = true;
vec[0].erase(0,1);
}
//Do not filter out any molecules if there is a parameter "showall";
//allows -s option to be used for highlighting substructures (--highlight also does this)
vector<string>::iterator it = std::remove(vec.begin(), vec.end(),"showall");
showAll = it != vec.end();
if(showAll)
vec.erase(it);
//Store the number of matches required, if as a number in the second parameter, else 0.
nmatches = 0;
comparechar = '\0';
if(vec.size()>1)
{
comparechar = vec[1][0];
if(comparechar=='>' || comparechar=='<')
vec[1].erase(0,1);
else
comparechar = '\0';
nmatches = atoi(vec[1].c_str());
if(nmatches) //remove this parameter to still allow coloring
vec.erase(vec.begin()+1);
}
//Interpret as a filename if possible
MakeQueriesFromMolInFile(queries, vec[0], &nPatternAtoms, strstr(OptionText,"noH"));
vec.erase(remove(vec.begin(),vec.end(),"noH"),vec.end());//to prevent "noH2" being seen as a color
if(queries.empty())
{
//SMARTS supplied
// Explicit H in SMARTS requires explicit H in the molecule.
// Calling AddHydrogens() on a copy of the molecule is done in parsmart.cpp
// only when SMARTS contains [H]. Doing more has complications with atom typing,
// so AddHydrogens here on the molecule (not a copy) when #1 detected.
addHydrogens = (vec[0].find("#1]")!=string::npos);
// If extra target mols have been supplied, make a composite SMARTS
// to test for any of the targets.
if(ExtraMols.size()>0)
{
for(unsigned i=0;i<ExtraMols.size();++i)
{
OBConversion extraConv;
extraConv.AddOption("h");
if(!extraConv.SetOutFormat("smi"))
return false;
// Add option which avoids implicit H being added to the SMARTS.
// The parameter must be present but can be anything.
extraConv.AddOption("h",OBConversion::OUTOPTIONS, "X");
xsmarts += ",$(" + extraConv.WriteString(ExtraMols[i], true) + ")";
}
}
string ysmarts = xsmarts.empty() ? vec[0] : "[$(" + vec[0] + ")" + xsmarts +"]";
xsmarts.clear();
if(!sp.Init(ysmarts))
{
string msg = ysmarts + " cannot be interpreted as either valid SMARTS "
"or the name of a file with an extension known to OpenBabel "
"that contains one or more pattern molecules.";
obErrorLog.ThrowError(__FUNCTION__, msg, obError, onceOnly);
delete pmol;
pmol = NULL;
pConv->SetOneObjectOnly(); //stop conversion
return false;
}
}
else
{
// Target is in a file. Add extra targets if any supplied
for(unsigned i=0;i<ExtraMols.size();++i)
queries.push_back(CompileMoleculeQuery(static_cast<OBMol*>(ExtraMols[i])));
ExtraMols.clear();
}
if(vec.size()>1 && vec[1]=="exact")
{
if(queries.empty())
{
//Convert SMARTS to SMILES to count number of atoms
OBConversion conv;
OBMol patmol;
if(!conv.SetInFormat("smi") || !conv.ReadString(&patmol, vec[0]))
{
obErrorLog.ThrowError(__FUNCTION__, "Cannot read the parameter of -s option, "
"which has to be valid SMILES when the exact option is used.", obError, onceOnly);
delete pmol;
if(pConv)
pConv->SetOneObjectOnly(); //stop conversion
return false;
}
nPatternAtoms = patmol.NumHvyAtoms();
}
}
else
nPatternAtoms = 0;
//disable old versions
if(pConv)
pConv->AddOption(GetID(), OBConversion::GENOPTIONS, "");
}
bool match = false;
//These are a vector of each mapping, each containing atom indxs.
vector<vector<int> > vecatomvec;
vector<vector<int> >* pMappedAtoms = NULL;
if(nPatternAtoms)
if(pmol->NumHvyAtoms() != nPatternAtoms)
return false;
unsigned int imol=0; //index of mol in pattern file
if(!queries.empty()) //filename supplied
{
//match is set true if any of the structures match - OR behaviour
for(qiter=queries.begin();qiter!=queries.end();++qiter, ++imol)
{
OBIsomorphismMapper* mapper = OBIsomorphismMapper::GetInstance(*qiter);
OBIsomorphismMapper::Mappings mappings;
mapper->MapUnique(pmol, mappings);
if( (match = !mappings.empty()) ) // extra parens to indicate truth value
{
OBIsomorphismMapper::Mappings::iterator ita;
OBIsomorphismMapper::Mapping::iterator itb;
for(ita=mappings.begin(); ita!=mappings.end();++ita)//each mapping
{
vector<int> atomvec;
for(itb=ita->begin(); itb!=ita->end();++itb)//each atom index
atomvec.push_back(itb->second+1);
vecatomvec.push_back(atomvec);
atomvec.clear();
}
pMappedAtoms = &vecatomvec;
break;
}
}
}
else //SMARTS supplied
{
if(addHydrogens)
pmol->AddHydrogens(false,false);
if( (match = sp.Match(*pmol)) ) // extra parens to indicate truth value
{
pMappedAtoms = &sp.GetMapList();
if(nmatches!=0)
{
int n = sp.GetUMapList().size();
if(comparechar=='>') match = (n > nmatches);
else if(comparechar=='<') match = (n < nmatches);
else match = (n == nmatches);
}
}
}
if((!showAll && (!match && !inv)) || (match && inv))
{
//delete a non-matching mol
delete pmol;
pmol = NULL;
return false;
}
if(match)
//Copy the idxes of the first match to a member variable so that it can be retrieved from outside
firstmatch.assign(pMappedAtoms->begin()->begin(), pMappedAtoms->begin()->end());
else
firstmatch.clear();
if(match && !inv && vec.size()>=2 && !vec[1].empty() && !nPatternAtoms)
{
vector<vector<int> >::iterator iter;
if (vec[1]=="extract" || (vec.size()>3 && vec[2]=="extract"))
{
//Delete all unmatched atoms. Use only the first match
ExtractSubstruct(pmol, *pMappedAtoms->begin());
return true;
}
// color the substructure if there is a second parameter which is not "exact" or "extract" or "noH"
// with multiple color parameters use the one corresponding to the query molecule, or the last
if(imol>vec.size()-2)
imol = vec.size()-2;
for(iter=pMappedAtoms->begin();iter!=pMappedAtoms->end();++iter)//each match
AddDataToSubstruct(pmol, *iter, "color", vec[imol+1]);
return true;
}
if(pConv && pConv->IsLast())
{
for(qiter=queries.begin();qiter!=queries.end();++qiter)
delete *qiter;
queries.clear();
}
return true;
}
int main(int argc,char *argv[])
{
// turn off slow sync with C-style output (we don't use it anyway).
std::ios::sync_with_stdio(false);
OBConversion conv;
OBFormat *inFormat, *canFormat;
OBMol mol;
ifstream ifs;
vector<OBMol> fragments;
unsigned int fragmentCount = 0; // track how many in library -- give a running count
map<string, int> index; // index of cansmi
string currentCAN;
unsigned int size;
OBAtom *atom;
OBBond *bond;
bool nonRingAtoms, nonRingBonds;
char buffer[BUFF_SIZE];
canFormat = conv.FindFormat("can");
conv.SetOutFormat(canFormat);
if (argc < 2)
{
cout << "Usage: obfragment <file>" << endl;
return(-1);
}
for (int i = 1; i < argc; i++) {
cerr << " Reading file " << argv[i] << endl;
inFormat = conv.FormatFromExt(argv[i]);
if(inFormat==NULL || !conv.SetInFormat(inFormat))
{
cerr << " Cannot read file format for " << argv[i] << endl;
continue; // try next file
}
ifs.open(argv[i]);
if (!ifs)
{
cerr << "Cannot read input file: " << argv[i] << endl;
continue;
}
while(ifs.peek() != EOF && ifs.good())
{
conv.Read(&mol, &ifs);
if (!mol.Has3D()) continue; // invalid coordinates!
mol.DeleteHydrogens(); // remove these before we do anything else
do {
nonRingAtoms = false;
size = mol.NumAtoms();
for (unsigned int i = 1; i <= size; ++i)
{
atom = mol.GetAtom(i);
if (!atom->IsInRing()) {
mol.DeleteAtom(atom);
nonRingAtoms = true;
break; // don't know how many atoms there are
}
// Previously, we changed atoms to carbon here.
// Now we perform this alchemy in terms of string-rewriting
// once the canonical SMILES is generated
}
} while (nonRingAtoms);
if (mol.NumAtoms() < 3)
continue;
if (mol.NumBonds() == 0)
continue;
do {
nonRingBonds = false;
size = mol.NumBonds();
for (unsigned int i = 0; i < size; ++i)
{
bond = mol.GetBond(i);
if (!bond->IsInRing()) {
mol.DeleteBond(bond);
nonRingBonds = true;
break; // don't know how many bonds there are
}
}
} while (nonRingBonds);
fragments = mol.Separate();
for (unsigned int i = 0; i < fragments.size(); ++i)
{
if (fragments[i].NumAtoms() < 3) // too small to care
continue;
currentCAN = conv.WriteString(&fragments[i], true);
currentCAN = RewriteSMILES(currentCAN); // change elements to "a/A" for compression
if (index.find(currentCAN) != index.end()) { // already got this
index[currentCAN] += 1; // add to the count for bookkeeping
continue;
}
index[currentCAN] = 1; // don't ever write this ring fragment again
// OK, now retrieve the canonical ordering for the fragment
vector<string> canonical_order;
if (fragments[i].HasData("Canonical Atom Order")) {
OBPairData *data = (OBPairData*)fragments[i].GetData("Canonical Atom Order");
tokenize(canonical_order, data->GetValue().c_str());
}
// Write out an XYZ-style file with the CANSMI as the title
cout << fragments[i].NumAtoms() << '\n';
cout << currentCAN << '\n'; // endl causes a flush
vector<string>::iterator can_iter;
unsigned int order;
OBAtom *atom;
fragments[i].Center();
fragments[i].ToInertialFrame();
for (unsigned int index = 0; index < canonical_order.size();
++index) {
order = atoi(canonical_order[index].c_str());
atom = fragments[i].GetAtom(order);
snprintf(buffer, BUFF_SIZE, "C%8.3f%8.3f%8.3f\n",
atom->x(), atom->y(), atom->z());
cout << buffer;
}
}
fragments.clear();
if (index.size() > fragmentCount) {
fragmentCount = index.size();
cerr << " Fragments: " << fragmentCount << endl;
}
} // while reading molecules (in this file)
ifs.close();
ifs.clear();
} // while reading files
// loop through the map and output frequencies
map<string, int>::const_iterator indexItr;
for (indexItr = index.begin(); indexItr != index.end(); ++indexItr) {
cerr << (*indexItr).second << " INDEX " << (*indexItr).first << "\n";
}
return(0);
}
void genericSmilesCanonicalTest(const std::string &smiles)
{
cout << "Testing generic smiles <-> canonical smiles" << endl;
// read a smiles string
OBMol mol;
OBConversion conv;
OB_REQUIRE( conv.SetInFormat("smi") );
OB_REQUIRE( conv.SetOutFormat("can") );
cout << "smiles: " << smiles << endl;
// read a smiles string
OB_REQUIRE( conv.ReadString(&mol, smiles) );
// store the stereo data for the smiles string using unique symmetry ids
std::vector<OBTetrahedralStereo::Config> tetrahedral1;
std::vector<OBCisTransStereo::Config> cistrans1;
std::vector<OBSquarePlanarStereo::Config> squareplanar1;
// get the stereo data
OB_ASSERT( mol.HasData(OBGenericDataType::StereoData) );
std::vector<OBGenericData *> stereoData = mol.GetAllData(OBGenericDataType::StereoData);
std::vector<unsigned int> canlbls;
std::vector<unsigned int> symclasses;
OBGraphSym gs1(&mol);
gs1.GetSymmetry(symclasses);
CanonicalLabels(&mol, symclasses, canlbls);
cout << "mol.NumAtoms = " << mol.NumAtoms() << endl;
for (std::vector<OBGenericData*>::iterator data = stereoData.begin(); data != stereoData.end(); ++data) {
if (((OBStereoBase*)*data)->GetType() == OBStereo::Tetrahedral) {
// convert to tetrahedral data
OBTetrahedralStereo *ts = dynamic_cast<OBTetrahedralStereo*>(*data);
OB_REQUIRE( ts );
OB_ASSERT( ts->IsValid() );
if (!ts->IsValid())
continue;
OBTetrahedralStereo::Config config = ts->GetConfig();
// convert atom ids to symmetry ids
if (mol.GetAtomById(config.center))
config.center = canlbls.at( mol.GetAtomById(config.center)->GetIdx() - 1 );
if (mol.GetAtomById(config.from))
config.from = canlbls.at( mol.GetAtomById(config.from)->GetIdx() - 1 );
if (mol.GetAtomById(config.refs[0]))
config.refs[0] = canlbls.at( mol.GetAtomById(config.refs[0])->GetIdx() - 1 );
if (mol.GetAtomById(config.refs[1]))
config.refs[1] = canlbls.at( mol.GetAtomById(config.refs[1])->GetIdx() - 1 );
if (mol.GetAtomById(config.refs[2]))
config.refs[2] = canlbls.at( mol.GetAtomById(config.refs[2])->GetIdx() - 1 );
cout << "Config with symmetry ids: " << config << endl;
tetrahedral1.push_back(config);
} else
if (((OBStereoBase*)*data)->GetType() == OBStereo::CisTrans) {
// convert to tetrahedral data
OBCisTransStereo *ct = dynamic_cast<OBCisTransStereo*>(*data);
OB_REQUIRE( ct );
OB_ASSERT( ct->IsValid() );
OBCisTransStereo::Config config = ct->GetConfig();
// convert atom ids to symmetry ids
config.begin = canlbls.at( mol.GetAtomById(config.begin)->GetIdx() - 1 );
config.end = canlbls.at( mol.GetAtomById(config.end)->GetIdx() - 1 );
if (mol.GetAtomById(config.refs[0]))
config.refs[0] = canlbls.at( mol.GetAtomById(config.refs[0])->GetIdx() - 1 );
if (mol.GetAtomById(config.refs[1]))
config.refs[1] = canlbls.at( mol.GetAtomById(config.refs[1])->GetIdx() - 1 );
if (mol.GetAtomById(config.refs[2]))
config.refs[2] = canlbls.at( mol.GetAtomById(config.refs[2])->GetIdx() - 1 );
if (mol.GetAtomById(config.refs[3]))
config.refs[3] = canlbls.at( mol.GetAtomById(config.refs[3])->GetIdx() - 1 );
cout << "Config with symmetry ids: " << config << endl;
cistrans1.push_back(config);
} else
if (((OBStereoBase*)*data)->GetType() == OBStereo::SquarePlanar) {
// convert to tetrahedral data
OBSquarePlanarStereo *sp = dynamic_cast<OBSquarePlanarStereo*>(*data);
OB_REQUIRE( sp );
OB_ASSERT( sp->IsValid() );
if (!sp->IsValid())
continue;
OBSquarePlanarStereo::Config config = sp->GetConfig();
// convert atom ids to symmetry ids
if (mol.GetAtomById(config.center))
config.center = canlbls.at( mol.GetAtomById(config.center)->GetIdx() - 1 );
if (mol.GetAtomById(config.refs[0]))
config.refs[0] = canlbls.at( mol.GetAtomById(config.refs[0])->GetIdx() - 1 );
if (mol.GetAtomById(config.refs[1]))
config.refs[1] = canlbls.at( mol.GetAtomById(config.refs[1])->GetIdx() - 1 );
if (mol.GetAtomById(config.refs[2]))
config.refs[2] = canlbls.at( mol.GetAtomById(config.refs[2])->GetIdx() - 1 );
if (mol.GetAtomById(config.refs[3]))
config.refs[3] = canlbls.at( mol.GetAtomById(config.refs[3])->GetIdx() - 1 );
cout << "Config with symmetry ids: " << config << endl;
squareplanar1.push_back(config);
}
}
// write to can smiles
std::string canSmiles = conv.WriteString(&mol);
cout << "canSmiles: " << canSmiles;
// read can smiles in again
OB_REQUIRE( conv.ReadString(&mol, canSmiles) );
// store the stereo data for the smiles string using unique symmetry ids
std::vector<OBTetrahedralStereo::Config> tetrahedral2;
std::vector<OBCisTransStereo::Config> cistrans2;
std::vector<OBSquarePlanarStereo::Config> squareplanar2;
// get the stereo data
OB_ASSERT( mol.HasData(OBGenericDataType::StereoData) );
stereoData = mol.GetAllData(OBGenericDataType::StereoData);
OBGraphSym gs2(&mol);
gs2.GetSymmetry(symclasses);
CanonicalLabels(&mol, symclasses, canlbls);
cout << "mol.NumAtoms = " << mol.NumAtoms() << endl;
for (std::vector<OBGenericData*>::iterator data = stereoData.begin(); data != stereoData.end(); ++data) {
if (((OBStereoBase*)*data)->GetType() == OBStereo::Tetrahedral) {
// convert to tetrahedral data
OBTetrahedralStereo *ts = dynamic_cast<OBTetrahedralStereo*>(*data);
OB_REQUIRE( ts );
OB_ASSERT( ts->IsValid() );
OBTetrahedralStereo::Config config = ts->GetConfig();
// convert atom ids to symmetry ids
if (mol.GetAtomById(config.center))
config.center = canlbls.at( mol.GetAtomById(config.center)->GetIdx() - 1 );
if (mol.GetAtomById(config.from))
config.from = canlbls.at( mol.GetAtomById(config.from)->GetIdx() - 1 );
if (mol.GetAtomById(config.refs[0]))
config.refs[0] = canlbls.at( mol.GetAtomById(config.refs[0])->GetIdx() - 1 );
if (mol.GetAtomById(config.refs[1]))
config.refs[1] = canlbls.at( mol.GetAtomById(config.refs[1])->GetIdx() - 1 );
if (mol.GetAtomById(config.refs[2]))
config.refs[2] = canlbls.at( mol.GetAtomById(config.refs[2])->GetIdx() - 1 );
cout << "Config with symmetry ids: " << config << endl;
tetrahedral2.push_back(config);
}
if (((OBStereoBase*)*data)->GetType() == OBStereo::CisTrans) {
// convert to tetrahedral data
OBCisTransStereo *ct = dynamic_cast<OBCisTransStereo*>(*data);
OB_REQUIRE( ct );
OB_ASSERT( ct->IsValid() );
OBCisTransStereo::Config config = ct->GetConfig();
// convert atom ids to symmetry ids
config.begin = canlbls.at( mol.GetAtomById(config.begin)->GetIdx() - 1 );
config.end = canlbls.at( mol.GetAtomById(config.end)->GetIdx() - 1 );
if (mol.GetAtomById(config.refs[0]))
config.refs[0] = canlbls.at( mol.GetAtomById(config.refs[0])->GetIdx() - 1 );
if (mol.GetAtomById(config.refs[1]))
config.refs[1] = canlbls.at( mol.GetAtomById(config.refs[1])->GetIdx() - 1 );
if (mol.GetAtomById(config.refs[2]))
config.refs[2] = canlbls.at( mol.GetAtomById(config.refs[2])->GetIdx() - 1 );
if (mol.GetAtomById(config.refs[3]))
config.refs[3] = canlbls.at( mol.GetAtomById(config.refs[3])->GetIdx() - 1 );
cout << "Config with symmetry ids: " << config << endl;
cistrans2.push_back(config);
} else
if (((OBStereoBase*)*data)->GetType() == OBStereo::SquarePlanar) {
// convert to tetrahedral data
OBSquarePlanarStereo *sp = dynamic_cast<OBSquarePlanarStereo*>(*data);
OB_REQUIRE( sp );
OB_ASSERT( sp->IsValid() );
OBSquarePlanarStereo::Config config = sp->GetConfig();
// convert atom ids to symmetry ids
if (mol.GetAtomById(config.center))
config.center = canlbls.at( mol.GetAtomById(config.center)->GetIdx() - 1 );
if (mol.GetAtomById(config.refs[0]))
config.refs[0] = canlbls.at( mol.GetAtomById(config.refs[0])->GetIdx() - 1 );
if (mol.GetAtomById(config.refs[1]))
config.refs[1] = canlbls.at( mol.GetAtomById(config.refs[1])->GetIdx() - 1 );
if (mol.GetAtomById(config.refs[2]))
config.refs[2] = canlbls.at( mol.GetAtomById(config.refs[2])->GetIdx() - 1 );
if (mol.GetAtomById(config.refs[3]))
config.refs[3] = canlbls.at( mol.GetAtomById(config.refs[3])->GetIdx() - 1 );
cout << "Config with symmetry ids: " << config << endl;
squareplanar2.push_back(config);
}
}
// compare the tetrahedral structs
OB_ASSERT( tetrahedral1.size() == tetrahedral2.size() );
for (unsigned int i = 0; i < tetrahedral1.size(); ++i) {
for (unsigned int j = 0; j < tetrahedral2.size(); ++j) {
if (tetrahedral1[i].center == tetrahedral2[j].center)
OB_ASSERT( tetrahedral1[i] == tetrahedral2[j] );
if ( tetrahedral1[i] != tetrahedral2[j] ) {
cout << "1 = " << tetrahedral1[i] << endl;
cout << "2 = " << tetrahedral2[j] << endl;
}
}
}
// compare the cistrans structs
OB_ASSERT( cistrans1.size() == cistrans2.size() );
for (unsigned int i = 0; i < cistrans1.size(); ++i) {
for (unsigned int j = 0; j < cistrans2.size(); ++j) {
if ((cistrans1[i].begin == cistrans2[j].begin) && (cistrans1[i].end == cistrans2[j].end))
OB_ASSERT( cistrans1[i] == cistrans2[j] );
if ((cistrans1[i].begin == cistrans2[j].end) && (cistrans1[i].end == cistrans2[j].begin))
OB_ASSERT( cistrans1[i] == cistrans2[j] );
}
}
// compare the square-planar structs
OB_ASSERT( squareplanar1.size() == squareplanar2.size() );
for (unsigned int i = 0; i < squareplanar1.size(); ++i) {
for (unsigned int j = 0; j < squareplanar2.size(); ++j) {
if (squareplanar1[i].center == squareplanar2[j].center)
OB_ASSERT( squareplanar1[i] == squareplanar2[j] );
if ( squareplanar1[i] != squareplanar2[j] ) {
cout << "1 = " << squareplanar1[i] << endl;
cout << "2 = " << squareplanar2[j] << endl;
}
}
}
cout << "." << endl << endl;
}
int main(int argc,char **argv)
{
char *program_name= argv[0];
int c;
char *FileIn = NULL;
if (argc != 2)
{
string err = "Usage: ";
err += program_name;
err += " <filename>\n"
"Output format:\n"
"name NAME\n"
"formula FORMULA\n"
"mol_weight MOLECULAR_WEIGHT\n"
"exact_mass ISOTOPIC MASS\n"
"canonical_SMILES STRING\n"
"InChI STRING\n"
"num_atoms NUM\n"
"num_bonds NUM\n"
"num_residues NUM\n"
"num_rotors NUM\n"
"sequence RESIDUE_SEQUENCE\n"
"num_rings NUMBER_OF_RING_(SSSR)\n"
"logP NUM\n"
"PSA POLAR_SURFACE_AREA\n"
"MR MOLAR REFRACTIVITY";
err += "$$$$";
// ThrowError(err); wasn't being output because error level too low
cerr << err; //Why not do directly
exit(-1);
}
else
{
FileIn = argv[1];
}
// Find Input filetype
OBConversion conv;
OBFormat *format = conv.FormatFromExt(FileIn);
if (!format || !conv.SetInFormat(format))
{
cerr << program_name << ": cannot read input format!" << endl;
exit (-1);
}
ifstream ifs;
// Read the file
ifs.open(FileIn);
if (!ifs)
{
cerr << program_name << ": cannot read input file!" << endl;
exit (-1);
}
OBMol mol;
OBFormat *canSMIFormat = conv.FindFormat("can");
OBFormat *inchiFormat = conv.FindFormat("inchi");
////////////////////////////////////////////////////////////////////////////
// List of properties
// Name
// Molecular weight (Standard molar mass given by IUPAC atomic masses)
// Number of rings : the size of the smallest set of smallest rings (SSSR)
//.....ADD YOURS HERE.....
for (c = 1;; ++c)
{
mol.Clear();
conv.Read(&mol, &ifs);
if (mol.Empty())
break;
if (!mol.HasHydrogensAdded())
mol.AddHydrogens();
// Print the properties
if (strlen(mol.GetTitle()) != 0)
cout << "name " << mol.GetTitle() << endl;
else
cout << "name " << FileIn << " " << c << endl;
cout << "formula " << mol.GetFormula() << endl;
cout << "mol_weight " << mol.GetMolWt() << endl;
cout << "exact_mass " << mol.GetExactMass() << endl;
string smilesString = "-";
if (canSMIFormat) {
conv.SetOutFormat(canSMIFormat);
smilesString = conv.WriteString(&mol);
if ( smilesString.length() == 0 )
{
smilesString = "-";
}
}
cout << "canonical_SMILES " << smilesString << endl;
string inchiString = "-";
if (inchiFormat) {
conv.SetOutFormat(inchiFormat);
inchiString = conv.WriteString(&mol);
if ( inchiString.length() == 0 )
{
inchiString = "-";
}
}
cout << "InChI " << inchiString << endl;
cout << "num_atoms " << mol.NumAtoms() << endl;
cout << "num_bonds " << mol.NumBonds() << endl;
cout << "num_residues " << mol.NumResidues() << endl;
cout << "num_rotors " << mol.NumRotors() << endl;
if (mol.NumResidues() > 0)
cout << "sequence " << sequence(mol) << endl;
else
cout << "sequence " << "-" << endl;
cout << "num_rings " << nrings(mol) << endl;
OBDescriptor* pDesc;
pDesc= OBDescriptor::FindType("logP");
if(pDesc)
cout << "logP " << pDesc->Predict(&mol) << endl;
pDesc = OBDescriptor::FindType("TPSA");
if(pDesc)
cout << "PSA " << pDesc->Predict(&mol) << endl;
pDesc = OBDescriptor::FindType("MR");
if(pDesc)
cout << "MR " << pDesc->Predict(&mol) << endl;
cout << "$$$$" << endl; // SDF like end of compound descriptor list
//Other OBDescriptors could be output here, even ones that were rarely
// used. Since these are plugin classes, they may not be loaded, but
// then with code like the above they are just ignored.
} // end for loop
return(0);
}
