// Helper function -- handle SMARTS selections
// Called by performAction()
void SelectExtension::selectSMARTS(GLWidget *widget)
{
bool ok;
QString pattern = QInputDialog::getText(qobject_cast<QWidget*>(parent()),
tr("SMARTS Selection"),
tr("SMARTS pattern to select"),
QLineEdit::Normal,
"", &ok);
if (ok && !pattern.isEmpty()) {
OBSmartsPattern smarts;
smarts.Init(pattern.toStdString());
OpenBabel::OBMol obmol = m_molecule->OBMol();
smarts.Match(obmol);
// if we have matches, select them
if(smarts.NumMatches() != 0) {
QList<Primitive *> matchedAtoms;
vector< vector <int> > mapList = smarts.GetUMapList();
vector< vector <int> >::iterator i; // a set of matching atoms
vector<int>::iterator j; // atom ids in each match
for (i = mapList.begin(); i != mapList.end(); ++i) {
for (j = i->begin(); j != i->end(); ++j) {
matchedAtoms.append(m_molecule->atom(obmol.GetAtom(*j)->GetIdx()-1));
}
}
widget->clearSelected();
widget->setSelected(matchedAtoms, true);
widget->update();
} // end matches
}
return;
}
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;
}
double Predict(OBBase* pOb, string* param=NULL)
{
OBMol* pmol = dynamic_cast<OBMol*> (pOb);
if(!pmol)
return 0;
OBSmartsPattern sp;
if (sp.Init(_smarts) && sp.Match(*pmol))
return sp.GetUMapList().size();
else
return 0.0;
}
bool GetFingerprint(OBBase* pOb, vector<unsigned int>&fp, int nbits)
{
OBMol* pmol = dynamic_cast<OBMol*>(pOb);
unsigned int o=0;
unsigned int m=0;
unsigned int i=0;
unsigned int n=0;
if(!pmol)
return false;
//Read patterns file if it has not been done already
if(smartsStrings.empty())
ReadPatternFile(_patternsfile, smartsStrings);
//Make fp size the smallest power of two to contain the patterns
//unsigned int n=Getbitsperint();
//while(n<smartsStrings.size())n*=2;
//fp.resize(n/Getbitsperint());
fp.resize(16);
for(n=0;n<smartsStrings.size();++n)
{
OBSmartsPattern sp;
sp.Init(smartsStrings[n]);
if(sp.Match(*pmol)) {
m=sp.GetUMapList().size();
//m=sp.NumMatches();
o=n*8;
for(i=0;i<8;++i) {
if(i<m) {SetBit(fp, o+i);
//cout << "1";
}
//cout << endl;
}
}
}
if(nbits)
Fold(fp, nbits);
return true;
};
///////////////////////////////////////////////////////////////////////////////
//! \brief Find the molecule(s) with or without a given SMART pattern
int main(int argc,char **argv)
{
char c;
unsigned int ntimes=0; // number of times SMARTS matches in a molecule
unsigned int numMatching = 0; // number of matching molecules (for -c flag)
bool pattern_matched=false, ntimes_matched=true;
bool count=false, invert=false, full=false, name_only=false;
char *FileIn = NULL, *Pattern = NULL;
char *program_name = argv[0];
char *iext;
bool useInFile = true;
OBConversion conv(&cin,&cout);
OBFormat *pFormat = conv.FindFormat("smi"); // default format is SMILES
// Parse options
while ((c = getopt(argc, argv, "t:nvcfi:-")) != -1)
{
#ifdef _WIN32
char optopt = c;
#endif
switch (c)
{
case 't': // request ntimes unique matches
c = sscanf(optarg, "%d", &ntimes);
if (c != 1 )
{
cerr << program_name << ": unable to parse -t option" << endl;
exit (-1);
}
break;
case 'i':
iext = optarg;
//The ID provided by the OBFormat class is used as
// the identifying file extension. This is a slight
// reduction in flexibility (which is not currently used)
pFormat = conv.FindFormat(iext);
if(pFormat==NULL)
{
cerr << program_name << ": cannot read input format!" << endl;
exit(-1);
}
break;
case 'n': // print the molecule name only
name_only = true;
break;
case 'c': // count the number of match
count = true;
break;
case 'v': // match only the molecules without the pattern
invert = true;
break;
case 'f':
full = true;
break;
case '-':
useInFile = false;
break;
case '?':
if (isprint (optopt))
fprintf (stderr, "Unknown option `-%c'.\n", optopt);
else
fprintf (stderr,
"Unknown option character `\\x%x'.\n",
optopt);
return 1;
}
}
int index = optind;
if (argc-index != 2 && argc-index != 1)
{
string err = "Usage: ";
err += program_name;
err += " [options] \"PATTERN\" <filename>\n";
err += "If no filename is supplied, then obgrep will use stdin instead.\n";
err += "Options:\n";
err += " -v Invert the matching, print non-matching molecules\n";
err += " -c Print the number of matched molecules\n";
err += " -i <format> Specify the input and output format\n";
err += " -f Full match, print matching-molecules when the number\n";
err += " of heavy atoms is equal to the number of PATTERN atoms\n";
err += " -n Only print the name of the molecules\n";
err += " -t NUM Print a molecule only if the PATTERN occurs NUM times inside the molecule\n";
cerr << err << ends;
exit(-1);
}
else
{
Pattern = argv[index++];
if (argc - index == 1)
FileIn = argv[index];
}
ifstream ifs;
if (useInFile && FileIn != NULL)
{
// Read the file
ifs.open(FileIn);
if (!ifs)
{
cerr << program_name << ": cannot read input file!" << endl;
exit (-1);
}
conv.SetInStream(&ifs);
// Find Input filetype
if (pFormat == NULL) {
pFormat = conv.FormatFromExt(FileIn);
if (pFormat == NULL)
{
cerr << program_name << ": cannot read input format!" << endl;
return (-1);
}
}
}
if (! conv.SetInAndOutFormats(pFormat, pFormat))
{
cerr << program_name << ": cannot read or write to this file format" << endl;
return (-1);
}
// Match the SMART
OBSmartsPattern sp;
vector< vector <int> > maplist; // list of matched atoms
sp.Init(Pattern);
OBMol mol;
bool impossible_match;
// Search for pattern
for (c=0;;)
{
mol.Clear();
conv.Read(&mol);
if (mol.Empty())
break;
////////////////////////////////////////////////////////////////
// Do not loose time trying to match the pattern if the matching
// is impossible.
// It is impossible to make a full match if the number of atoms is
// different
if (full )
impossible_match = (sp.NumAtoms() == mol.NumHvyAtoms()) ? false : true;
else
impossible_match = false;
if (impossible_match)
{ // -> avoid useless SMART matching attempt
if (invert)
{
if (!count)
{
if ( name_only )
cout << mol.GetTitle() << endl;
else
conv.Write(&mol, &cout);
}
numMatching++;
}
continue;
}
////////////////////////////////////////////////////////////////
// perform SMART matching
pattern_matched = sp.Match(mol);
// the number of times the match occured may matter
if ( ntimes )
{ // ntimes is a positive integer of requested matches
// Here, a match mean a unique match (same set of atoms)
// so we need to get the unique match list size
maplist = sp.GetUMapList();
if( maplist.size() == ntimes )
ntimes_matched = true;
else
ntimes_matched = false;
}
else
{ // ntimes == 0, we don't care about the number of matches
ntimes_matched = true;
}
////////////////////////////////////////////////////////////////
// perform a set of tests to guess what to print out
if ( pattern_matched == true && ntimes_matched == true)
{
if (!invert)
{ // do something only when invert flag is off
if (!count)
{
if ( name_only )
cout << mol.GetTitle() << endl;
else
conv.Write(&mol, &cout);
}
numMatching++;
}
}
else
{ // The SMART pattern do not occur as many times as requested
if (invert)
{ // do something only if invert flag is on
if (!count)
{
if ( name_only )
cout << mol.GetTitle() << endl;
else
conv.Write(&mol, &cout);
}
numMatching++;
}
}
} // end for loop
////////////////////////////////////////////////////////////////
// Only print the number of matched molecules as requested
if (count)
{
cout << numMatching << endl;
}
return(1);
}
///////////////////////////////////////////////////////////////////////////////
//! \brief Set a tortional bond to a given angle
int main(int argc,char **argv)
{
const char *Pattern=NULL;
unsigned int i, t, errflg = 0;
int c;
char flags[255];
string err;
bool graphOutput=false;
// parse the command line -- optional -a flag to change all matching torsions
if (argc < 3 || argc > 4) {
errflg++;
} else {
FileIn = argv[1];
Pattern = "[!$(*#*)&!D1]-!@[!$(*#*)&!D1]";
// Read the atom position
c = sscanf(argv[2], "%d", &angleSum);
angle = 360./angleSum;
if (argc == 4)
{
c = sscanf(argv[3], "%s", flags);
int flagid=1;
while (flags[flagid]!=0)
switch (flags[flagid++])
{
case 'g':
graphOutput=true;
case 'e':
forceField=OBForceField::FindForceField("MMFF94");
isEnergyCalcing=true;
break;
}
}
}
if (errflg) {
cerr << "Usage: rkrotate <filename> <angle> [options]" << endl;
exit(-1);
}
// create pattern
OBSmartsPattern sp;
sp.Init(Pattern);
OBFormat* format = conv.FormatFromExt(FileIn);
if(!(format && conv.SetInAndOutFormats(format, format))) { //in and out formats same
cerr << "obrotate: cannot read and/or write this file format!" << endl;
exit (-1);
} //...NF
//Open the molecule file
ifstream ifs;
// Read the file
ifs.open(FileIn);
if (!ifs) {
cerr << "obrotate: cannot read input file!" << endl;
exit (-1);
}
OBMol mol;
vector< vector <int> > maplist; // list of matched atoms
// vector< vector <int> >::iterator m; // and its iterators
// int tindex;
// Set the angles
for (;;) {
mol.Clear();
//NF ifs >> mol; // Read molecule
conv.Read(&mol,&ifs); //NF
if (mol.Empty())
break;
if (sp.Match(mol)) {
// if match perform rotation
maplist = sp.GetUMapList(); // get unique matches
if (maplist.size() > 1)
cerr << "obrotate: Found " << maplist.size() << " matches." << endl;
energySheet=new MultiVector<double>(degrees=maplist.size(),angleSum);
indexSheet=new int[maplist.size()];
for (int EXO=0;EXO<maplist.size();++EXO)
totalSum*=angleSum+EXO;
// look at all the mapping atom but save only the last one.
turnMol(mol,maplist,maplist.size()-1);
if (graphOutput)
{
ofstream ofs("energyGraph.mlog");
int ind[degrees];
for (int i=0;i<degrees;++i)
ind[i]=0;
do
{
for (int i=0;i<degrees;++i)
ofs<<ind[i]<<'\t';
ofs<<energySheet->getVectorValue(ind)<<endl;
}
while(energySheet->incressIndex(ind));
}
if (isEnergyCalcing)
{
std::vector<int*> lowEnergySheet;
totalSum=energySheet->getMinValues(lowEnergySheet);
if (totalSum)
outputMol(lowEnergySheet,mol,maplist,maplist.size()-1);
else
cerr << "rkrotate: No low energy conformation found." << endl;
}
cout << sum;
} else {
cerr << "obrotate: Found 0 matches for the SMARTS pattern." << endl;
exit(-1);
}
//NF cout << mol;
}
return(0);
}
void OBBondTyper::AssignFunctionalGroupBonds(OBMol &mol)
{
if (!_init)
Init();
OBSmartsPattern *currentPattern;
OBBond *b1, *b2;
OBAtom *a1,*a2, *a3;
double angle, dist1, dist2;
vector<int> assignments;
vector<vector<int> > mlist;
vector<vector<int> >::iterator matches, l;
vector<pair<OBSmartsPattern*, vector<int> > >::iterator i;
unsigned int j;
// Loop through for all the functional groups and assign bond orders
for (i = _fgbonds.begin();i != _fgbonds.end();++i)
{
currentPattern = i->first;
assignments = i->second;
if (currentPattern && currentPattern->Match(mol))
{
mlist = currentPattern->GetUMapList();
for (matches = mlist.begin(); matches != mlist.end(); ++matches)
{
// Now loop through the bonds to assign from _fgbonds
for (j = 0; j < assignments.size(); j += 3)
{
// along the assignments vector: atomID1 atomID2 bondOrder
a1 = mol.GetAtom((*matches)[ assignments[j] ]);
a2 = mol.GetAtom((*matches)[ assignments[j+1 ] ]);
if (!a1 || !a2) continue;
b1 = a1->GetBond(a2);
if (!b1) continue;
b1->SetBO(assignments[j+2]);
} // bond order assignments
} // each match
} // current pattern matches
} // for(functional groups)
// FG with distance and/or bond criteria
// Carbonyl oxygen C=O
OBSmartsPattern carbo; carbo.Init("[#8D1][#6](*)(*)");
if (carbo.Match(mol))
{
mlist = carbo.GetUMapList();
for (l = mlist.begin(); l != mlist.end(); ++l)
{
a1 = mol.GetAtom((*l)[0]);
a2 = mol.GetAtom((*l)[1]);
angle = a2->AverageBondAngle();
dist1 = a1->GetDistance(a2);
// carbonyl geometries ?
if (angle > 115 && angle < 150 && dist1 < 1.28) {
if ( !a1->HasDoubleBond() ) {// no double bond already assigned
b1 = a1->GetBond(a2);
if (!b1 ) continue;
b1->SetBO(2);
}
}
}
} // Carbonyl oxygen
// thione C=S
OBSmartsPattern thione; thione.Init("[#16D1][#6](*)(*)");
if (thione.Match(mol))
{
mlist = thione.GetUMapList();
for (l = mlist.begin(); l != mlist.end(); ++l)
{
a1 = mol.GetAtom((*l)[0]);
a2 = mol.GetAtom((*l)[1]);
angle = a2->AverageBondAngle();
dist1 = a1->GetDistance(a2);
// thione geometries ?
if (angle > 115 && angle < 150 && dist1 < 1.72) {
if ( !a1->HasDoubleBond() ) {// no double bond already assigned
b1 = a1->GetBond(a2);
if (!b1 ) continue;
b1->SetBO(2);
}
}
}
} // thione
// Isocyanate N=C=O or Isothiocyanate
bool dist1OK;
OBSmartsPattern isocyanate; isocyanate.Init("[#8,#16;D1][#6D2][#7D2]");
if (isocyanate.Match(mol))
{
mlist = isocyanate.GetUMapList();
for (l = mlist.begin(); l != mlist.end(); ++l)
{
a1 = mol.GetAtom((*l)[0]);
a2 = mol.GetAtom((*l)[1]);
a3 = mol.GetAtom((*l)[2]);
angle = a2->AverageBondAngle();
dist1 = a1->GetDistance(a2);
dist2 = a2->GetDistance(a3);
// isocyanate geometry or Isotiocyanate geometry ?
if (a1->IsOxygen())
dist1OK = dist1 < 1.28;
else
dist1OK = dist1 < 1.72;
if (angle > 150 && dist1OK && dist2 < 1.34) {
b1 = a1->GetBond(a2);
b2 = a2->GetBond(a3);
if (!b1 || !b2) continue;
b1->SetBO(2);
b2->SetBO(2);
}
}
} // Isocyanate
// oxime C=S
OBSmartsPattern oxime; oxime.Init("[#6D3][#7D2][#8D2]");
if (oxime.Match(mol))
{
mlist = oxime.GetUMapList();
for (l = mlist.begin(); l != mlist.end(); ++l)
{
a1 = mol.GetAtom((*l)[0]);
a2 = mol.GetAtom((*l)[1]);
angle = a2->AverageBondAngle();
dist1 = a1->GetDistance(a2);
// thione geometries ?
if (angle > 110 && angle < 150 && dist1 < 1.4) {
if ( !a1->HasDoubleBond() ) {// no double bond already assigned
b1 = a1->GetBond(a2);
if (!b1 ) continue;
b1->SetBO(2);
}
}
}
} // oxime
}