void testTetrahedralStereo1()
{
cout << "testTetrahedralStereo1()" << endl;
// read a smiles string
OBMol mol;
OBConversion conv;
OB_REQUIRE( conv.SetInFormat("smi") );
cout << "smiles: C[C@H](O)N" << endl;
OB_REQUIRE( conv.ReadString(&mol, "C[C@H](O)N") );
// get the stereo data
OB_REQUIRE( mol.HasData(OBGenericDataType::StereoData) );
std::vector<OBGenericData *> stereoData = mol.GetAllData(OBGenericDataType::StereoData);
OB_REQUIRE( stereoData.size() == 1 );
// convert to tetrahedral data
OB_REQUIRE( ((OBStereoBase*)stereoData[0])->GetType() == OBStereo::Tetrahedral );
OBTetrahedralStereo *ts = dynamic_cast<OBTetrahedralStereo*>(stereoData[0]);
OB_REQUIRE( ts );
// print the configuration
cout << *ts << endl;
// construct a valid configuration here
//
// C[C@H](O)N
// 0 1 2 3 4 <- ids
//
OBTetrahedralStereo::Config cfg(1, 0, OBStereo::MakeRefs(4, 3, 2), OBStereo::Clockwise);
// compare stereochemistry
OB_REQUIRE( ts->GetConfig() == cfg );
cout << endl;
}
void testIsomorphism1()
{
OBMol mol;
OBConversion conv;
conv.SetInFormat("smi");
conv.ReadString(&mol, "CC1CCC(C)CC1");
OBQuery *query = CompileMoleculeQuery(&mol);
OBIsomorphismMapper *mapper = OBIsomorphismMapper::GetInstance(query);
OBIsomorphismMapper::Mappings maps;
mapper->MapAll(&mol, maps);
OB_ASSERT( maps.size() == 4 );
delete query;
delete mapper;
query = CompileSmilesQuery("C1(C)CCC(C)CC1");
mapper = OBIsomorphismMapper::GetInstance(query);
OBIsomorphismMapper::Mapping map;
mapper->MapFirst(&mol, map);
OB_ASSERT( map.size() == 8 );
mapper->MapUnique(&mol, maps);
OB_ASSERT( maps.size() == 1 );
mapper->MapAll(&mol, maps);
OB_ASSERT( maps.size() == 4 );
delete query;
delete mapper;
}
void CheckValidDipeptide(OBConversion &conv,
const string &test,
unsigned int testCount)
{
OBMol mol;
OBResidue *res;
ostringstream os;
mol.Clear();
conv.ReadString(&mol, test);
chainsparser.PerceiveChains(mol);
if (mol.NumResidues() != 2) {
os << "not ok " << testCount << " # expected 2 residues, but found "
<< mol.NumResidues() << '\n';
os << "# ";
FOR_RESIDUES_OF_MOL(res, mol)
os << res->GetName() << " ";
os << endl;
BOOST_CHECK_MESSAGE( 0, os.str().c_str() );
} else {
res = mol.GetResidue(0);
BOOST_CHECK_MESSAGE( res, "Get first AA from dipeptide" );
res = mol.GetResidue(1);
BOOST_CHECK_MESSAGE( res, "Get second AA from dipeptide" );
}
}
// Delete hydrogens should not remove charged or isotopic hydrogens or [H][H] or [Cu][H][Cu]
// or hydrogens with assigned atom classes
void test_Issue178_DeleteHydrogens()
{
OBConversion conv;
conv.SetInFormat("smi");
OBMol mol;
// Test DeleteHydrogens() and DeleteNonPolarHydrogens()
static const char *smi[] = { "C[H]", "[H][H]", "C[1H]", "C[H]C", "C[H+]" };
int numHs[] = { 0, 2, 1, 1, 1 };
for (int i = 0; i < 5; ++i) {
for (int j = 0; j < 2; ++j) {
conv.ReadString(&mol, smi[i]);
if (j == 0)
mol.DeleteHydrogens();
else
mol.DeleteNonPolarHydrogens();
int myNumHs = 0;
FOR_ATOMS_OF_MOL(atom, mol)
if (atom->IsHydrogen())
myNumHs++;
OB_COMPARE(myNumHs, numHs[i]);
}
}
// Test DeletePolarHydrogens()
static const char *smiB[] = { "N[H]", "[H][H]", "N[1H]", "N[H]C", "N[H+]" };
int numHsB[] = { 0, 2, 1, 1, 1 };
for (int i = 0; i < 5; ++i) {
conv.ReadString(&mol, smiB[i]);
mol.DeletePolarHydrogens();
int myNumHs = 0;
FOR_ATOMS_OF_MOL(atom, mol)
if (atom->IsHydrogen())
myNumHs++;
OB_COMPARE(myNumHs, numHsB[i]);
}
// Test atom class
// Currently, the SMILES parser does not retain atom classes for hydrogens on reading so...
conv.ReadString(&mol, "C[H]");
OBAtomClassData *ac = new OBAtomClassData;
ac->Add(2, 99); // Assign the hydrogen (atom 2) a class of 99
mol.SetData(ac);
mol.DeleteHydrogens();
int myNumHs = 0;
FOR_ATOMS_OF_MOL(atom, mol)
if (atom->IsHydrogen())
myNumHs++;
OB_COMPARE(myNumHs, 1);
}
// 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 testAutomorphismPreMapping()
{
cout << "testAutomorphismPreMapping" << endl;
OBMol mol;
OBConversion conv;
conv.SetInFormat("smi");
conv.ReadString(&mol, "c1(C)c(C)c(C)c(C)c(C)c1");
Automorphisms aut;
FindAutomorphisms((OBMol*)&mol, aut);
cout << aut.size() << endl;
OB_ASSERT( aut.size() == 2 );
}
// 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");
}
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;
}
void testIsomorphism4()
{
cout << "testIsomorphism4" << endl;
OBMol mol;
OBConversion conv;
conv.SetInFormat("smi");
conv.ReadString(&mol, "C12C(C2)C1");
OBQuery *query = CompileSmilesQuery("C1CC1");
OBIsomorphismMapper *mapper = OBIsomorphismMapper::GetInstance(query);
OBIsomorphismMapper::Mappings maps;
mapper->MapUnique(&mol, maps);
cout << maps.size() << endl;
OB_ASSERT( maps.size() == 2 );
delete query;
delete mapper;
}
int main()
{
// Create an OBConversion object.
OBConversion conv;
// Set the input format.
if (!conv.SetInFormat("smi")) {
// Handle error.
return 1;
}
// Create the OBMol object.
OBMol mol;
// Read the smiles string.
if (conv.ReadString(&mol, "CCCC")) {
// Handle error.
return 1;
}
// ...Use OBMol object...
}
void CheckInvalidResidue(OBConversion &conv,
const string &test,
unsigned int testCount)
{
OBMol mol;
mol.Clear();
conv.ReadString(&mol, test);
chainsparser.PerceiveChains(mol);
if (mol.NumResidues() != 0) {
OBResidue *res = mol.GetResidue(0);
if (res->GetName() == "LIG") { // ligand, not residue
cout << "ok " << testCount << " # found ligand, not residue "
<< test << '\n';
} else {
cout << "not ok " << testCount << " # expected 0 residues, found "
<< mol.NumResidues() << '\n';
cout << "# " << res->GetName() << endl;
}
} else
cout << "ok " << testCount << " # correctly rejected " << test << '\n';
}
bool OBReader::readSmiString(QString smiString)
{
using namespace OpenBabel;
OBConversion conv;
if (!conv.SetInFormat("smi"))
{
qDebug() << "Error in conv.SetInFormat().";
return false;
}
OBMol obMol;
if (!conv.ReadString(&obMol, smiString.toStdString()))
{
qDebug() << "Error occured while reading the smi string.";
return false;
}
if (!obMol.Has3D())
{
if (!buildGeometry(&obMol))
{
qDebug() << "Error in buildGeometry()";
return false;
}
}
if (!toMolecule(&obMol))
{
qDebug() << "Could not convert OBMol to Molecule.";
return false;
}
return true;
}
void CheckInvalidResidue(OBConversion &conv,
const string &test,
unsigned int testCount)
{
OBMol mol;
ostringstream os;
mol.Clear();
conv.ReadString(&mol, test);
chainsparser.PerceiveChains(mol);
if (mol.NumResidues() != 0) {
OBResidue *res = mol.GetResidue(0);
if (res->GetName() == "LIG") { // ligand, not residue
BOOST_CHECK( 1 );
} else {
os << "not ok " << testCount << " # expected 0 residues, found "
<< mol.NumResidues() << '\n';
os << "# " << res->GetName() << endl;
BOOST_CHECK_MESSAGE( 0, os.str().c_str() );
}
} else
BOOST_CHECK( 1 );
}
void CheckValidDipeptide(OBConversion &conv,
const string &test,
unsigned int testCount)
{
OBMol mol;
mol.Clear();
conv.ReadString(&mol, test);
chainsparser.PerceiveChains(mol);
if (mol.NumResidues() != 2) {
cout << "not ok " << testCount << " # expected 2 residues, but found "
<< mol.NumResidues() << '\n';
cout << "# ";
FOR_RESIDUES_OF_MOL(res, mol)
cout << res->GetName() << " ";
cout << endl;
} else {
OBResidue *res;
res = mol.GetResidue(0);
cout << "ok " << testCount << " # " << res->GetName();
res = mol.GetResidue(1);
cout << " " << res->GetName() << '\n';
}
}
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;
}
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;
}
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 static variables vec, inv. The parameter is cleared so that:
// (a) the original -s option in transform.cpp is inactive, and
// (b) the parsing does not have to be done again for multi-molecule files
string txt(pmap->find(GetID())->second); // ID can be "s" or "v"
static vector<string> vec;
static bool inv;
static int nPatternAtoms; //non-zero for exact matches
static OBQuery* query;
static vector<OBQuery*> queries;
vector<OBQuery*>::iterator qiter;
if(!txt.empty())
{
//Set up on first call
tokenize(vec, txt);
inv = GetID()[0]=='v';
if(vec[0][0]=='~')
{
inv = true;
vec[0].erase(0,1);
}
//Interpret as a filename if possible
MakeQueriesFromMolInFile(queries, vec[0], &nPatternAtoms);
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;
pConv->SetOneObjectOnly(); //stop conversion
return false;
}
nPatternAtoms = patmol.NumHvyAtoms();
}
}
else
nPatternAtoms = 0;
//disable old versions
pConv->AddOption(GetID(), OBConversion::GENOPTIONS, "");
}
bool match;
//These are a vector of each mapping, each containing atom indxs.
vector<vector<int> > vecatomvec;
vector<vector<int> >* pMappedAtoms = NULL;
OBSmartsPattern sp;
if(nPatternAtoms)
if(pmol->NumHvyAtoms() != nPatternAtoms)
return false;
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(!sp.Init(vec[0]))
{
string msg = vec[0] + " 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;
}
if( (match = sp.Match(*pmol)) ) // extra parens to indicate truth value
pMappedAtoms = &sp.GetMapList();
}
if((!match && !inv) || (match && inv))
{
//delete a non-matching mol
delete pmol;
pmol = NULL;
return false;
}
if(!inv && vec.size()>=2 && !vec[1].empty() && !nPatternAtoms)
{
vector<vector<int> >::iterator iter;
if(vec[1]=="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"
// 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;
}
void residue_test()
{
cout << "# Unit tests for OBResidue \n";
// OBResidue isolation tests
OBResidue emptyResidue, testRes1;
// chains parser tests
// PR#1515198
static const string loopTest1("C1(C(NC(C(N1C(C(NC(C=Cc1ccccc1)=O)C)=O)Cc1ccccc1)=O)Cc1ccccc1)=O");
OBConversion conv;
OBMol mol;
OBFormat *inFormat = conv.FindFormat("SMI");
conv.SetInFormat(inFormat);
conv.ReadString(&mol, loopTest1);
chainsparser.PerceiveChains(mol);
// parse common residues
unsigned int testCount = 3;
static const string ala("NC(C)C(O)(=O)");
CheckValidResidue(conv, ala, ++testCount);
static const string arg("NC(CCCNC(N)=N)C(O)(=O)");
CheckValidResidue(conv, arg, ++testCount);
static const string asn("NC(CC(N)=O)C(O)(=O)");
CheckValidResidue(conv, asn, ++testCount);
static const string asp("NC(CC(O)=O)C(O)(=O)");
CheckValidResidue(conv, asp, ++testCount);
static const string cys("NC(CS)C(O)(=O)");
CheckValidResidue(conv, cys, ++testCount);
static const string glu("NC(CCC(O)=O)C(O)(=O)");
CheckValidResidue(conv, glu, ++testCount);
static const string gln("NC(CCC(N)=O)C(O)(=O)");
CheckValidResidue(conv, gln, ++testCount);
static const string gly("NC([H])C(O)(=O)");
CheckValidResidue(conv, gly, ++testCount);
static const string his("NC(CC1=CNC=N1)C(O)(=O)");
CheckValidResidue(conv, his, ++testCount);
static const string ile("NC(C(CC)C)C(O)(=O)");
CheckValidResidue(conv, ile, ++testCount);
static const string leu("NC(CC(C)C)C(O)(=O)");
CheckValidResidue(conv, leu, ++testCount);
static const string lys("NC(CCCCN)C(O)(=O)");
CheckValidResidue(conv, lys, ++testCount);
static const string met("NC(CCSC)C(O)(=O)");
CheckValidResidue(conv, met, ++testCount);
static const string phe("NC(CC1=CC=CC=C1)C(O)(=O)");
CheckValidResidue(conv, phe, ++testCount);
static const string pro("OC(C1CCCN1)(=O)");
CheckValidResidue(conv, pro, ++testCount);
static const string ser("NC(CO)C(O)(=O)");
CheckValidResidue(conv, ser, ++testCount);
static const string thr("NC(C(C)O)C(O)(=O)");
CheckValidResidue(conv, thr, ++testCount);
static const string trp("NC(CC1=CNC2=C1C=CC=C2)C(O)(=O)");
CheckValidResidue(conv, trp, ++testCount);
static const string tyr("NC(CC1=CC=C(O)C=C1)C(O)(=O)");
CheckValidResidue(conv, tyr, ++testCount);
static const string val("NC(C(C)C)C(O)(=O)");
CheckValidResidue(conv, val, ++testCount);
// nucleics
static const string a("OC[C@H]1O[C@H](C[C@@H]1O)n1cnc2c(ncnc12)N");
CheckValidResidue(conv, a, ++testCount);
static const string g("OC[C@H]1O[C@H](C[C@@H]1O)n1c(nc(cc1)N)=O");
CheckValidResidue(conv, g, ++testCount);
static const string c("OC[C@H]1O[C@H](C[C@@H]1O)n1cnc2c([nH]c(nc12)N)=O");
CheckValidResidue(conv, c, ++testCount);
static const string t("OC[C@H]1O[C@H](C[C@@H]1O)n1c([nH]c(c(c1)C)=O)=O");
CheckValidResidue(conv, t, ++testCount);
static const string u("OC[C@H]1O[C@H]([C@@H]([C@@H]1O)O)n1c([nH]c(cc1)=O)=O");
CheckValidResidue(conv, u, ++testCount);
// invalid residues
static const string benzene("c1ccccc1");
CheckInvalidResidue(conv, benzene, ++testCount);
static const string pyrrole("c1cccn[H]1");
CheckInvalidResidue(conv, pyrrole, ++testCount);
static const string amine("CC(=O)CCN");
CheckInvalidResidue(conv, amine, ++testCount);
// check some dipeptides
static const string ala_val("NC(C)C(=O)NC(C(C)C)C(=O)O");
CheckValidDipeptide(conv, ala_val, ++testCount);
static const string cys_leu("NC(CS)C(=O)NC(CC(C)C)C(=O)O");
CheckValidDipeptide(conv, cys_leu, ++testCount);
}
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);
}
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[])
{
// 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: residue" << endl;
cout << " Unit tests for OBResidue " << endl;
return(-1);
}
cout << "# Unit tests for OBResidue \n";
cout << "ok 1\n"; // for loading tests
// OBResidue isolation tests
OBResidue emptyResidue, testRes1;
cout << "ok 2\n"; // ctor works
// chains parser tests
// PR#1515198
static const string loopTest1("C1(C(NC(C(N1C(C(NC(C=Cc1ccccc1)=O)C)=O)Cc1ccccc1)=O)Cc1ccccc1)=O");
OBConversion conv;
OBMol mol;
OBFormat *inFormat = conv.FindFormat("SMI");
conv.SetInFormat(inFormat);
conv.ReadString(&mol, loopTest1);
chainsparser.PerceiveChains(mol);
// OK if it doesn't crash
cout << "ok 3\n";
// parse common residues
unsigned int testCount = 3;
static const string ala("NC(C)C(O)(=O)");
CheckValidResidue(conv, ala, ++testCount);
static const string arg("NC(CCCNC(N)=N)C(O)(=O)");
CheckValidResidue(conv, arg, ++testCount);
static const string asn("NC(CC(N)=O)C(O)(=O)");
CheckValidResidue(conv, asn, ++testCount);
static const string asp("NC(CC(O)=O)C(O)(=O)");
CheckValidResidue(conv, asp, ++testCount);
static const string cys("NC(CS)C(O)(=O)");
CheckValidResidue(conv, cys, ++testCount);
static const string glu("NC(CCC(O)=O)C(O)(=O)");
CheckValidResidue(conv, glu, ++testCount);
static const string gln("NC(CCC(N)=O)C(O)(=O)");
CheckValidResidue(conv, gln, ++testCount);
static const string gly("NC([H])C(O)(=O)");
CheckValidResidue(conv, gly, ++testCount);
static const string his("NC(CC1=CNC=N1)C(O)(=O)");
CheckValidResidue(conv, his, ++testCount);
static const string ile("NC(C(CC)C)C(O)(=O)");
CheckValidResidue(conv, ile, ++testCount);
static const string leu("NC(CC(C)C)C(O)(=O)");
CheckValidResidue(conv, leu, ++testCount);
static const string lys("NC(CCCCN)C(O)(=O)");
CheckValidResidue(conv, lys, ++testCount);
static const string met("NC(CCSC)C(O)(=O)");
CheckValidResidue(conv, met, ++testCount);
static const string phe("NC(CC1=CC=CC=C1)C(O)(=O)");
CheckValidResidue(conv, phe, ++testCount);
static const string pro("OC(C1CCCN1)(=O)");
CheckValidResidue(conv, pro, ++testCount);
static const string ser("NC(CO)C(O)(=O)");
CheckValidResidue(conv, ser, ++testCount);
static const string thr("NC(C(C)O)C(O)(=O)");
CheckValidResidue(conv, thr, ++testCount);
static const string trp("NC(CC1=CNC2=C1C=CC=C2)C(O)(=O)");
CheckValidResidue(conv, trp, ++testCount);
static const string tyr("NC(CC1=CC=C(O)C=C1)C(O)(=O)");
CheckValidResidue(conv, tyr, ++testCount);
static const string val("NC(C(C)C)C(O)(=O)");
CheckValidResidue(conv, val, ++testCount);
// nucleics
static const string a("OC[C@H]1O[C@H](C[C@@H]1O)n1cnc2c(ncnc12)N");
CheckValidResidue(conv, a, ++testCount);
static const string g("OC[C@H]1O[C@H](C[C@@H]1O)n1c(nc(cc1)N)=O");
CheckValidResidue(conv, g, ++testCount);
static const string c("OC[C@H]1O[C@H](C[C@@H]1O)n1cnc2c([nH]c(nc12)N)=O");
CheckValidResidue(conv, c, ++testCount);
static const string t("OC[C@H]1O[C@H](C[C@@H]1O)n1c([nH]c(c(c1)C)=O)=O");
CheckValidResidue(conv, t, ++testCount);
static const string u("OC[C@H]1O[C@H]([C@@H]([C@@H]1O)O)n1c([nH]c(cc1)=O)=O");
CheckValidResidue(conv, u, ++testCount);
// invalid residues
static const string benzene("c1ccccc1");
CheckInvalidResidue(conv, benzene, ++testCount);
static const string pyrrole("c1cccn[H]1");
CheckInvalidResidue(conv, pyrrole, ++testCount);
static const string amine("CC(=O)CCN");
CheckInvalidResidue(conv, amine, ++testCount);
// check some dipeptides
static const string ala_val("NC(C)C(=O)NC(C(C)C)C(=O)O");
CheckValidDipeptide(conv, ala_val, ++testCount);
static const string cys_leu("NC(CS)C(=O)NC(CC(C)C)C(=O)O");
CheckValidDipeptide(conv, cys_leu, ++testCount);
// the number of tests for "prove"
cout << "1.." << testCount << "\n";
return(0);
}
void phmodel_test()
{
OBMol mol;
OBConversion conv;
conv.SetInFormat("smi");
conv.SetOutFormat("smi");
unsigned int test = 0;
// amine acid COOH pKa = 4.0 (carboxylic acid entry in phmodel.txt)
// amino acid NH3+ pKa = 10.0 (amine entry in phmodel.txt)
//
// Aspartic acid (sidechain COOH pKa = 3.8)
//
conv.ReadString(&mol, "NC(CC(O)=O)C(O)=O");
mol.SetAutomaticFormalCharge(true);
mol.AddHydrogens(false, true, 1.0); // NH3+ COOH COOH
//conv.Write(&mol, &cout);
BOOST_CHECK_MESSAGE( mol.NumAtoms() == 17, "Aspartic acid pH 1.0" );
conv.ReadString(&mol, "NC(CC(O)=O)C(O)=O");
mol.SetAutomaticFormalCharge(true);
mol.AddHydrogens(false, true, 3.9); // NH3+ COOH COO-
//conv.Write(&mol, &cout);
BOOST_CHECK_MESSAGE( mol.NumAtoms() == 16, "Aspartic acid pH 3.9" );
conv.ReadString(&mol, "NC(CC(O)=O)C(O)=O");
mol.SetAutomaticFormalCharge(true);
mol.AddHydrogens(false, true, 7.4); // NH3+ COO- COO-
//conv.Write(&mol, &cout);
BOOST_CHECK_MESSAGE( mol.NumAtoms() == 15, "Aspartic acid pH 7.4" );
conv.ReadString(&mol, "NC(CC(O)=O)C(O)=O");
mol.SetAutomaticFormalCharge(true);
mol.AddHydrogens(false, true, 13.0); // NH2 COO- COO-
//conv.Write(&mol, &cout);
BOOST_CHECK_MESSAGE( mol.NumAtoms() == 14, "Aspartic acid pH 13.0" );
//
// Glutamic acid (sidechain COOH pKa = 4.3)
//
conv.ReadString(&mol, "NC(CCC(O)=O)C(O)=O");
mol.SetAutomaticFormalCharge(true);
mol.AddHydrogens(false, true, 1.0); // NH3+ COOH COOH
//conv.Write(&mol, &cout);
BOOST_CHECK_MESSAGE( mol.NumAtoms() == 20, "Glutamic acid pH 1.0" );
conv.ReadString(&mol, "NC(CCC(O)=O)C(O)=O");
mol.SetAutomaticFormalCharge(true);
mol.AddHydrogens(false, true, 4.15); // NH3+ COOH COO-
//conv.Write(&mol, &cout);
// known bug
BOOST_CHECK_MESSAGE( mol.NumAtoms() == 19, "Glutamic acid pH 4.15" );
conv.ReadString(&mol, "NC(CCC(O)=O)C(O)=O");
mol.SetAutomaticFormalCharge(true);
mol.AddHydrogens(false, true, 7.4); // NH3+ COO- COO-
//conv.Write(&mol, &cout);
BOOST_CHECK_MESSAGE( mol.NumAtoms() == 18, "Glutamic acid pH 7.4" );
conv.ReadString(&mol, "NC(CCC(O)=O)C(O)=O");
mol.SetAutomaticFormalCharge(true);
mol.AddHydrogens(false, true, 13.0); // NH2 COO- COO-
//conv.Write(&mol, &cout);
BOOST_CHECK_MESSAGE( mol.NumAtoms() == 17, "Glutamic acid pH 13.0" );
//
// Histidine (sidechain nH+ pKa = 6.08)
//
conv.ReadString(&mol, "NC(Cc1ncnc1)C(O)=O");
mol.SetAutomaticFormalCharge(true);
mol.AddHydrogens(false, true, 1.0); // NH3+ COOH nH+
//conv.Write(&mol, &cout);
BOOST_CHECK_MESSAGE( mol.NumAtoms() == 22, "Histidine pH 1.0" );
conv.ReadString(&mol, "NC(Cc1ncnc1)C(O)=O");
mol.SetAutomaticFormalCharge(true);
mol.AddHydrogens(false, true, 5.0); // NH3+ COO- nH+
//conv.Write(&mol, &cout);
BOOST_CHECK_MESSAGE( mol.NumAtoms() == 21, "Histidine pH 5.0" );
conv.ReadString(&mol, "NC(Cc1ncnc1)C(O)=O");
mol.SetAutomaticFormalCharge(true);
mol.AddHydrogens(false, true, 7.4); // NH3+ COO- n:
//conv.Write(&mol, &cout);
BOOST_CHECK_MESSAGE( mol.NumAtoms() == 20, "Histidine pH 7.4" );
conv.ReadString(&mol, "NC(Cc1ncnc1)C(O)=O");
mol.SetAutomaticFormalCharge(true);
mol.AddHydrogens(false, true, 13.0); // NH2 COO- n:
//conv.Write(&mol, &cout);
BOOST_CHECK_MESSAGE( mol.NumAtoms() == 19, "Histidine pH 13.0" );
//
// Lysine (sidechain NH3+ pKa = 8.28)
//
conv.ReadString(&mol, "NC(CCCCN)C(O)=O");
mol.SetAutomaticFormalCharge(true);
mol.AddHydrogens(false, true, 1.0); // NH3+ COOH NH3+
//conv.Write(&mol, &cout);
BOOST_CHECK_MESSAGE( mol.NumAtoms() == 26, "Lysine pH 1.0" );
conv.ReadString(&mol, "NC(CCCCN)C(O)=O");
mol.SetAutomaticFormalCharge(true);
mol.AddHydrogens(false, true, 7.4); // NH3+ COO- NH3+
//conv.Write(&mol, &cout);
BOOST_CHECK_MESSAGE( mol.NumAtoms() == 25, "Lysine pH 7.4" );
conv.ReadString(&mol, "NC(CCCCN)C(O)=O");
mol.SetAutomaticFormalCharge(true);
mol.AddHydrogens(false, true, 9.0); // NH3+ COO- NH2
//conv.Write(&mol, &cout);
// known bug
BOOST_CHECK_MESSAGE( mol.NumAtoms() == 24, "Lysine pH 9.0" );
conv.ReadString(&mol, "NC(CCCCN)C(O)=O");
mol.SetAutomaticFormalCharge(true);
mol.AddHydrogens(false, true, 13.0); // NH2 COO- NH2
//conv.Write(&mol, &cout);
BOOST_CHECK_MESSAGE( mol.NumAtoms() == 23, "Lysine pH 13.0" );
//
// Tyrosine (sidechain OH pKa = 10.1)
//
conv.ReadString(&mol, "NC(Cc1ccc(O)cc1)C(O)=O");
mol.SetAutomaticFormalCharge(true);
mol.AddHydrogens(false, true, 1.0); // NH3+ COOH NH3+
//conv.Write(&mol, &cout);
BOOST_CHECK_MESSAGE( mol.NumAtoms() == 25, "Tyrosine pH 1.0" );
conv.ReadString(&mol, "NC(Cc1ccc(O)cc1)C(O)=O");
mol.SetAutomaticFormalCharge(true);
mol.AddHydrogens(false, true, 7.4); // NH3+ COO- NH3+
//conv.Write(&mol, &cout);
BOOST_CHECK_MESSAGE( mol.NumAtoms() == 24, "Tyrosine pH 7.4" );
conv.ReadString(&mol, "NC(Cc1ccc(O)cc1)C(O)=O");
mol.SetAutomaticFormalCharge(true);
mol.AddHydrogens(false, true, 10.05); // NH3+ COO- NH2
//conv.Write(&mol, &cout);
BOOST_CHECK_MESSAGE( mol.NumAtoms() == 23, "Tyrosine pH 10.05" );
conv.ReadString(&mol, "NC(Cc1ccc(O)cc1)C(O)=O");
mol.SetAutomaticFormalCharge(true);
mol.AddHydrogens(false, true, 13.0); // NH2 COO- NH2
//conv.Write(&mol, &cout);
BOOST_CHECK_MESSAGE( mol.NumAtoms() == 22, "Tyrosine pH 13.0" );
//
// Arginine (sidechain =NH2+ pKa = 12.0)
//
conv.ReadString(&mol, "NC(CCCNC(N)=N)C(O)=O");
mol.SetAutomaticFormalCharge(true);
mol.AddHydrogens(false, true, 1.0); // NH3+ COOH =NH2+
//conv.Write(&mol, &cout);
BOOST_CHECK_MESSAGE( mol.NumAtoms() == 28, "Arginine pH 1.0" );
conv.ReadString(&mol, "NC(CCCNC(N)=N)C(O)=O");
mol.SetAutomaticFormalCharge(true);
mol.AddHydrogens(false, true, 7.4); // NH3+ COO- NH3+
//conv.Write(&mol, &cout);
BOOST_CHECK_MESSAGE( mol.NumAtoms() == 27, "Arginine pH 7.4" );
conv.ReadString(&mol, "NC(CCCNC(N)=N)C(O)=O");
mol.SetAutomaticFormalCharge(true);
mol.AddHydrogens(false, true, 11.0); // NH3+ COO- NH2
//conv.Write(&mol, &cout);
BOOST_CHECK_MESSAGE( mol.NumAtoms() == 26, "Arginine pH 11.0" );
conv.ReadString(&mol, "NC(CCCNC(N)=N)C(O)=O");
mol.SetAutomaticFormalCharge(true);
mol.AddHydrogens(false, true, 13.0); // NH2 COO- NH2
//conv.Write(&mol, &cout);
BOOST_CHECK_MESSAGE( mol.NumAtoms() == 25, "Arginine pH 13.0" );
//
// Gly-Gly (terminal NH3+, COOH and the amide bond)
//
conv.ReadString(&mol, "NCC(=O)NCC(=O)O");
mol.SetAutomaticFormalCharge(true);
mol.AddHydrogens(false, true, 1.0); // NH3+ COOH =NH2+
//conv.Write(&mol, &cout);
BOOST_CHECK_MESSAGE( mol.NumAtoms() == 18, "Gly-Gly pH 1.0" );
conv.ReadString(&mol, "NCC(=O)NCC(=O)O");
mol.SetAutomaticFormalCharge(true);
mol.AddHydrogens(false, true, 7.4); // NH3+ COO- NH3+
//conv.Write(&mol, &cout);
BOOST_CHECK_MESSAGE( mol.NumAtoms() == 17, "Gly-Gly pH 7.4" );
conv.ReadString(&mol, "NCC(=O)NCC(=O)O");
mol.SetAutomaticFormalCharge(true);
mol.AddHydrogens(false, true, 13.0); // NH2 COO- NH2
//conv.Write(&mol, &cout);
BOOST_CHECK_MESSAGE( mol.NumAtoms() == 16, "Gly-Gly pH 13.0" );
}
void readSmiles(const std::string &smiles, OBMol &mol)
{
OBConversion conv;
conv.SetInFormat("smi");
conv.ReadString(&mol, smiles);
}