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);
}
bool MakeQueriesFromMolInFile(vector<OBQuery*>& queries, const std::string& filename, int* pnAtoms, bool noH)
{
OBMol patternMol;
patternMol.SetIsPatternStructure();
OBConversion patternConv;
OBFormat* pFormat;
//Need to distinguish between filename and SMARTS. Not infallable...
if( filename.empty() ||
filename.find('.')==string::npos ||
!(pFormat = patternConv.FormatFromExt(filename.c_str())) ||
!patternConv.SetInFormat(pFormat) ||
!patternConv.ReadFile(&patternMol, filename) ||
patternMol.NumAtoms()==0)
return false;
if(noH)
patternMol.DeleteHydrogens();
do
{
*pnAtoms = patternMol.NumHvyAtoms();
queries.push_back(CompileMoleculeQuery(&patternMol));
}while(patternConv.Read(&patternMol));
return true;
}
void LinMorphExtension::loadFile(QString file)
{
qDebug("LinMorphExtension::loadFile()");
if (file.isEmpty())
return;
if (!m_secondMolecule)
m_secondMolecule = new Molecule;
OBConversion conv;
OBFormat *inFormat = OBConversion::FormatFromExt(( file.toAscii() ).data() );
if ( !inFormat || !conv.SetInFormat( inFormat ) ) {
QMessageBox::warning( NULL, tr( "Avogadro" ),
tr( "Cannot read file format of file %1." )
.arg( file ) );
return;
}
if (!conv.ReadFile(m_secondMolecule, file.toStdString())) {
QMessageBox::warning( NULL, tr( "Avogadro" ),
tr( "Read mol file %1 failed." )
.arg( file ) );
return;
}
qDebug("LinMorphExtension::loadFile complete");
computeConformers(m_secondMolecule);
m_linMorphDialog->setFrame(1);
m_timeLine->setFrameRange(1, m_frameCount);
setDuration(m_linMorphDialog->fps());
}
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 ColoredMol::color()
{
OBConversion conv;
std::string ext = boost::filesystem::extension(ligName);
if(ext.compare(".pdb") == 0)
{
conv.SetInFormat("PDB");
}
else if(ext.compare(".pdbqt") == 0)
{
conv.SetInFormat("PDBQT");
}
else
{
std::cout << "File extension not supported: " << ligName << '\n';
std::cout << "Please use .pdb or .pdbqt for ligand\n";
exit(0);
}
conv.ReadFile(&ligMol, ligName);
ext = boost::filesystem::extension(recName);
if(ext.compare(".pdb") == 0)
{
conv.SetInFormat("PDB");
}
else
{
std::cout << "File extension not supported: " << recName << '\n';
std::cout << "Please use .pdb for receptor\n";
exit(0);
}
conv.ReadFile(&recMol, recName);
addHydrogens();
ligCenter();
removeResidues();
removeEachAtom();
}
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);
}
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);
}
void testAlternativeOrigin()
{
// See https://github.com/openbabel/openbabel/pull/1558
OBConversion conv;
OBMol mol;
conv.SetInFormat("cif");
conv.ReadFile(&mol, GetFilename("test04.cif"));
OBUnitCell* pUC = (OBUnitCell*)mol.GetData(OBGenericDataType::UnitCell);
const SpaceGroup* pSG = pUC->GetSpaceGroup();
SpaceGroup* sg = new SpaceGroup(*pSG);
pSG = SpaceGroup::Find(sg);
string summary = obErrorLog.GetMessageSummary();
OB_ASSERT( summary.find("warning") == string::npos);
OB_ASSERT( pSG != NULL );
OB_ASSERT( pSG->GetOriginAlternative() == 1);
}
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);
}
void testDecayToP1()
{
// See https://github.com/openbabel/openbabel/pull/261
OBConversion conv;
OBMol mol;
conv.SetInFormat("cif");
conv.ReadFile(&mol, GetFilename("test03.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("2 warnings") != string::npos);
OB_ASSERT( pSG->GetId() == 1 );
}
int main(int argc, char *argv[])
{
string prog_name(argv[0]);
string root;
string::size_type i = prog_name.find_last_of("/\\");
if(i == string::npos)
root = "./";
else
root = prog_name.substr(0,i+1);
if(argc<3 || argc>7) {
cerr << endl << " OBJ : to find maximum common subgraph between mol1 and mol2" << endl
<< endl << " Usage: " << argv[0] << "[options] mol1 mol2" << endl
<< " [options]" << endl
<< " --rmax n: stop recursion after 'n' steps" << endl
<< " (default: do not stop)" << endl
<< " --atom file: where atom types are defined" << endl
<< " (default: " << root << "atom_type.txt)" << endl << endl
<< " Attention: for molecules with lots of atoms, it could be very time-consuming!" << endl
<< " In that case, you can use './mcs' to get a quasi-MCS instead!" << endl << endl;
exit(EXIT_FAILURE);
}
int rmax = 0;
atom_type_file = root + "atom_type.txt";
int ii;
for(ii=1; ii<argc; ) {
if(strncmp(argv[ii],"--",2) != 0)
break;
if(strcmp(argv[ii],"--rmax") == 0)
rmax = atoi(argv[ii+1]);
else if(strcmp(argv[ii],"--atom") == 0)
atom_type_file = argv[ii+1];
else {
cerr << "Error: invalid option <" << argv[ii] << ">" << endl;
exit(EXIT_FAILURE);
}
ii += 2;
}
if(argc-ii != 2) {
cerr << "Error: invalid number of arguments, need 2 molecule files" << endl;
exit(EXIT_FAILURE);
}
string name1(argv[ii]);
string name2(argv[ii+1]);
string format1 = name1.substr(name1.rfind(".")+1);
string format2 = name2.substr(name2.rfind(".")+1);
OBConversion conv;
OBMol mol1,mol2;
if(!conv.SetInFormat(format1.c_str()) || !conv.ReadFile(&mol1,name1)) {
cerr << "Error: failed to read molecule from " << name1 << endl;
exit(EXIT_FAILURE);
}
if(!conv.SetInFormat(format2.c_str()) || !conv.ReadFile(&mol2,name2)) {
cerr << "Error: failed to read molecule from " << name2 << endl;
exit(EXIT_FAILURE);
}
vector<pair<vector<int>,vector<int> > > graph_indices = mcs(mol1,mol2,rmax);
for(vector<pair<vector<int>,vector<int> > >::iterator i=graph_indices.begin(); i!=graph_indices.end(); ++i) {
cout << endl << "atom mapping:" << endl;
cout << " g1:";
for(vector<int>::iterator j=(*i).first.begin();j!=(*i).first.end();++j)
printf(" %-2d",*j+1);
cout << endl << " g2:";
for(vector<int>::iterator j=(*i).second.begin();j!=(*i).second.end();++j)
printf(" %-2d",*j+1);
cout << endl;
}
cout << endl << "totally " << graph_indices.size() << " atom mappings for MCS" << endl << endl;
return 0;
}
int main(int argc, char* argv[])
{
// Check the required number of command line arguments.
if (argc != 5)
{
cout << "usr host user pwd jobs_path" << endl;
return 0;
}
// Fetch command line arguments.
const auto host = argv[1];
const auto user = argv[2];
const auto pwd = argv[3];
const path jobs_path = argv[4];
// Connect to host and authenticate user.
DBClientConnection conn;
{
cout << local_time() << "Connecting to " << host << " and authenticating " << user << endl;
string errmsg;
if ((!conn.connect(host, errmsg)) || (!conn.auth("istar", user, pwd, errmsg)))
{
cerr << local_time() << errmsg << endl;
return 1;
}
}
// Initialize constants.
cout << local_time() << "Initializing" << endl;
const auto collection = "istar.usr";
const auto epoch = date(1970, 1, 1);
const size_t num_usrs = 2;
constexpr array<size_t, num_usrs> qn{{ 12, 60 }};
constexpr array<double, num_usrs> qv{{ 1.0 / qn[0], 1.0 / qn[1] }};
const size_t num_references = 4;
const size_t num_subsets = 5;
const array<string, num_subsets> SubsetSMARTS
{{
"[!#1]", // heavy
"[#6+0!$(*~[#7,#8,F]),SH0+0v2,s+0,S^3,Cl+0,Br+0,I+0]", // hydrophobic
"[a]", // aromatic
"[$([O,S;H1;v2]-[!$(*=[O,N,P,S])]),$([O,S;H0;v2]),$([O,S;-]),$([N&v3;H1,H2]-[!$(*=[O,N,P,S])]),$([N;v3;H0]),$([n,o,s;+0]),F]", // acceptor
"[N!H0v3,N!H0+v4,OH+0,SH+0,nH+0]", // donor
}};
// Initialize variables.
array<array<double, qn.back()>, 1> qw;
array<array<double, qn.back()>, 1> lw;
auto q = qw[0];
auto l = lw[0];
// Read ZINC ID file.
const string_array<size_t> zincids("16_zincid.txt");
const auto num_ligands = zincids.size();
// Read SMILES file.
const string_array<size_t> smileses("16_smiles.txt");
assert(smileses.size() == num_ligands);
// Read supplier file.
const string_array<size_t> suppliers("16_supplier.txt");
assert(suppliers.size() == num_ligands);
// Read property files of floating point types and integer types.
const auto zfproperties = read<array<float, 4>>("16_zfprop.f32");
assert(zfproperties.size() == num_ligands);
const auto ziproperties = read<array<int16_t, 5>>("16_ziprop.i16");
assert(ziproperties.size() == num_ligands);
// Open files for subsequent reading.
std::ifstream usrcat_bin("16_usrcat.f64");
stream_array<size_t> ligands("16_ligand.pdbqt");
assert(ligands.size() == num_ligands);
array<vector<double>, 2> scores
{{
vector<double>(num_ligands, 0),
vector<double>(num_ligands, 0)
}};
const auto& u0scores = scores[0];
const auto& u1scores = scores[1];
vector<size_t> scase(num_ligands);
// Enter event loop.
cout << local_time() << "Entering event loop" << endl;
bool sleeping = false;
while (true)
{
// Fetch an incompleted job in a first-come-first-served manner.
if (!sleeping) cout << local_time() << "Fetching an incompleted job" << endl;
BSONObj info;
conn.runCommand("istar", BSON("findandmodify" << "usr" << "query" << BSON("done" << BSON("$exists" << false) << "started" << BSON("$exists" << false)) << "sort" << BSON("submitted" << 1) << "update" << BSON("$set" << BSON("started" << Date_t(duration_cast<std::chrono::milliseconds>(system_clock::now().time_since_epoch()).count())))), info); // conn.findAndModify() is available since MongoDB C++ Driver legacy-1.0.0
const auto value = info["value"];
if (value.isNull())
{
// No incompleted jobs. Sleep for a while.
if (!sleeping) cout << local_time() << "Sleeping" << endl;
sleeping = true;
this_thread::sleep_for(chrono::seconds(10));
continue;
}
sleeping = false;
const auto job = value.Obj();
// Obtain job properties.
const auto _id = job["_id"].OID();
cout << local_time() << "Executing job " << _id.str() << endl;
const auto job_path = jobs_path / _id.str();
const auto format = job["format"].String();
const auto email = job["email"].String();
// Parse the user-supplied ligand.
OBMol obMol;
OBConversion obConversion;
obConversion.SetInFormat(format.c_str());
obConversion.ReadFile(&obMol, (job_path / ("ligand." + format)).string());
const auto num_atoms = obMol.NumAtoms();
// obMol.AddHydrogens(); // Adding hydrogens does not seem to affect SMARTS matching.
// Classify subset atoms.
array<vector<int>, num_subsets> subsets;
for (size_t k = 0; k < num_subsets; ++k)
{
auto& subset = subsets[k];
subset.reserve(num_atoms);
OBSmartsPattern smarts;
smarts.Init(SubsetSMARTS[k]);
smarts.Match(obMol);
for (const auto& map : smarts.GetMapList())
{
subset.push_back(map.front());
}
}
const auto& subset0 = subsets.front();
// Check user-provided ligand validity.
if (subset0.empty())
{
// Record job completion time stamp.
const auto millis_since_epoch = duration_cast<std::chrono::milliseconds>(system_clock::now().time_since_epoch()).count();
conn.update(collection, BSON("_id" << _id), BSON("$set" << BSON("done" << Date_t(millis_since_epoch))));
// Send error notification email.
cout << local_time() << "Sending an error notification email to " << email << endl;
MailMessage message;
message.setSender("usr <*****@*****.**>");
message.setSubject("Your usr job has failed");
message.setContent("Description: " + job["description"].String() + "\nSubmitted: " + to_simple_string(ptime(epoch, boost::posix_time::milliseconds(job["submitted"].Date().millis))) + " UTC\nFailed: " + to_simple_string(ptime(epoch, boost::posix_time::milliseconds(millis_since_epoch))) + " UTC\nReason: failed to parse the provided ligand.");
message.addRecipient(MailRecipient(MailRecipient::PRIMARY_RECIPIENT, email));
SMTPClientSession session("137.189.91.190");
session.login();
session.sendMessage(message);
session.close();
continue;
}
// Calculate the four reference points.
const auto n = subset0.size();
const auto v = 1.0 / n;
array<vector3, num_references> references{};
auto& ctd = references[0];
auto& cst = references[1];
auto& fct = references[2];
auto& ftf = references[3];
for (const auto i : subset0)
{
ctd += obMol.GetAtom(i)->GetVector();
}
ctd *= v;
double cst_dist = numeric_limits<double>::max();
double fct_dist = numeric_limits<double>::lowest();
double ftf_dist = numeric_limits<double>::lowest();
for (const auto i : subset0)
{
const auto& a = obMol.GetAtom(i)->GetVector();
const auto this_dist = a.distSq(ctd);
if (this_dist < cst_dist)
{
cst = a;
cst_dist = this_dist;
}
if (this_dist > fct_dist)
{
fct = a;
fct_dist = this_dist;
}
}
for (const auto i : subset0)
{
const auto& a = obMol.GetAtom(i)->GetVector();
const auto this_dist = a.distSq(fct);
if (this_dist > ftf_dist)
{
ftf = a;
ftf_dist = this_dist;
}
}
// Precalculate the distances between each atom and each reference point.
array<vector<double>, num_references> dista;
for (size_t k = 0; k < num_references; ++k)
{
const auto& reference = references[k];
auto& dists = dista[k];
dists.resize(1 + num_atoms); // OpenBabel atom index starts from 1. dists[0] is dummy.
for (size_t i = 0; i < n; ++i)
{
dists[subset0[i]] = sqrt(obMol.GetAtom(subset0[i])->GetVector().distSq(reference));
}
}
// Calculate USR and USRCAT features of the input ligand.
size_t qo = 0;
for (const auto& subset : subsets)
{
const auto n = subset.size();
for (size_t k = 0; k < num_references; ++k)
{
const auto& distp = dista[k];
vector<double> dists(n);
for (size_t i = 0; i < n; ++i)
{
dists[i] = distp[subset[i]];
}
array<double, 3> m{};
if (n > 2)
{
const auto v = 1.0 / n;
for (size_t i = 0; i < n; ++i)
{
const auto d = dists[i];
m[0] += d;
}
m[0] *= v;
for (size_t i = 0; i < n; ++i)
{
const auto d = dists[i] - m[0];
m[1] += d * d;
}
m[1] = sqrt(m[1] * v);
for (size_t i = 0; i < n; ++i)
{
const auto d = dists[i] - m[0];
m[2] += d * d * d;
}
m[2] = cbrt(m[2] * v);
}
else if (n == 2)
{
m[0] = 0.5 * (dists[0] + dists[1]);
m[1] = 0.5 * fabs(dists[0] - dists[1]);
}
else if (n == 1)
{
m[0] = dists[0];
}
#pragma unroll
for (const auto e : m)
{
q[qo++] = e;
}
}
}
assert(qo == qn.back());
// Compute USR and USRCAT scores.
usrcat_bin.seekg(0);
for (size_t k = 0; k < num_ligands; ++k)
{
usrcat_bin.read(reinterpret_cast<char*>(l.data()), sizeof(l));
double s = 0;
#pragma unroll
for (size_t i = 0, u = 0; u < num_usrs; ++u)
{
#pragma unroll
for (const auto qnu = qn[u]; i < qnu; ++i)
{
s += fabs(q[i] - l[i]);
}
scores[u][k] = s;
}
}
assert(usrcat_bin.tellg() == sizeof(l) * num_ligands);
// Sort ligands by USRCAT score and then by USR score and then by ZINC ID.
iota(scase.begin(), scase.end(), 0);
sort(scase.begin(), scase.end(), [&](const size_t val0, const size_t val1)
{
const auto u1score0 = u1scores[val0];
const auto u1score1 = u1scores[val1];
if (u1score0 == u1score1)
{
const auto u0score0 = u0scores[val0];
const auto u0score1 = u0scores[val1];
if (u0score0 == u0score1)
{
return zincids[val0] < zincids[val1];
}
return u0score0 < u0score1;
}
return u1score0 < u1score1;
});
// Write results.
filtering_ostream log_csv_gz;
log_csv_gz.push(gzip_compressor());
log_csv_gz.push(file_sink((job_path / "log.csv.gz").string()));
log_csv_gz.setf(ios::fixed, ios::floatfield);
log_csv_gz << "ZINC ID,USR score,USRCAT score\n" << setprecision(8);
filtering_ostream ligands_pdbqt_gz;
ligands_pdbqt_gz.push(gzip_compressor());
ligands_pdbqt_gz.push(file_sink((job_path / "ligands.pdbqt.gz").string()));
ligands_pdbqt_gz.setf(ios::fixed, ios::floatfield);
for (size_t t = 0; t < 10000; ++t)
{
const size_t k = scase[t];
const auto zincid = zincids[k].substr(0, 8); // Take another substr() to get rid of the trailing newline.
const auto u0score = 1 / (1 + scores[0][k] * qv[0]);
const auto u1score = 1 / (1 + scores[1][k] * qv[1]);
log_csv_gz << zincid << ',' << u0score << ',' << u1score << '\n';
// Only write conformations of the top ligands to ligands.pdbqt.gz.
if (t >= 1000) continue;
const auto zfp = zfproperties[k];
const auto zip = ziproperties[k];
ligands_pdbqt_gz
<< "MODEL " << '\n'
<< "REMARK 911 " << zincid
<< setprecision(3)
<< ' ' << setw(8) << zfp[0]
<< ' ' << setw(8) << zfp[1]
<< ' ' << setw(8) << zfp[2]
<< ' ' << setw(8) << zfp[3]
<< ' ' << setw(3) << zip[0]
<< ' ' << setw(3) << zip[1]
<< ' ' << setw(3) << zip[2]
<< ' ' << setw(3) << zip[3]
<< ' ' << setw(3) << zip[4]
<< '\n'
<< "REMARK 912 " << smileses[k] // A newline is already included in smileses[k].
<< "REMARK 913 " << suppliers[k] // A newline is already included in suppliers[k].
<< setprecision(8)
<< "REMARK 951 USR SCORE: " << setw(10) << u0score << '\n'
<< "REMARK 952 USRCAT SCORE: " << setw(10) << u1score << '\n'
;
const auto lig = ligands[k];
ligands_pdbqt_gz.write(lig.data(), lig.size());
ligands_pdbqt_gz << "ENDMDL\n";
}
// Update progress.
cout << local_time() << "Setting done time" << endl;
const auto millis_since_epoch = duration_cast<std::chrono::milliseconds>(system_clock::now().time_since_epoch()).count();
conn.update(collection, BSON("_id" << _id), BSON("$set" << BSON("done" << Date_t(millis_since_epoch))));
// Send completion notification email.
cout << local_time() << "Sending a completion notification email to " << email << endl;
MailMessage message;
message.setSender("istar <*****@*****.**>");
message.setSubject("Your usr job has completed");
message.setContent("Description: " + job["description"].String() + "\nSubmitted: " + to_simple_string(ptime(epoch, boost::posix_time::milliseconds(job["submitted"].Date().millis))) + " UTC\nCompleted: " + to_simple_string(ptime(epoch, boost::posix_time::milliseconds(millis_since_epoch))) + " UTC\nResult: http://istar.cse.cuhk.edu.hk/usr/iview/?" + _id.str());
message.addRecipient(MailRecipient(MailRecipient::PRIMARY_RECIPIENT, email));
SMTPClientSession session("137.189.91.190");
session.login();
session.sendMessage(message);
session.close();
}
}
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);
}
extern "C" int readgrid_(float *vdata, int *nx, int *ny, int *nz,
float *x0, float *y0, float *z0, float *xx, float *yy, float *zz,
char *file, int fsize) {
float dx,dy,dz;
int gsize;
float v, vmin, vmax, vavg;
int navg = 0;
OBMol mol;
OBConversion conv;
conv.SetInFormat("cube");
std::string fname = file;
int blank = fname.find(" ");
//printf("%d,'%s'\n", blank, fname.substr(0,blank).c_str());
conv.ReadFile(&mol, fname.substr(0,blank).c_str());
//cout << mol.NumAtoms() << " atoms." << endl;
if (mol.HasData(OBGenericDataType::GridData)) {
std::vector<OBGenericData*> grids = mol.GetAllData(OBGenericDataType::GridData);
OBGridData *grid = dynamic_cast<OBGridData *> (grids[0]);
gsize = grid->GetNumberOfPoints();
grid->GetNumberOfPoints(*nx, *ny, *nz);
vector3 origin = grid->GetOriginVector();
*x0 = origin[0]; *y0 = origin[1]; *z0 = origin[2];
vector3 maxv = grid->GetMaxVector();
*xx = maxv[0]; *yy = maxv[1]; *zz = maxv[2];
dx=(*xx-*x0)/(*nx-1);
dy=(*yy-*y0)/(*ny-1);
dz=(*zz-*z0)/(*nz-1);
printf("%s %d=%d*%d*%d\n", grid->GetAttribute().c_str(), gsize, *nx, *ny, *nz);
printf("%f %f\n", grid->GetMinValue(), grid->GetMaxValue());
printf("%f,%f,%f\n", *x0,*y0,*z0);
printf("%f,%f,%f\n", *xx,*yy,*zz);
printf("%f,%f,%f\n", dx,dy,dz);
//vdata = (float *)calloc(gsize, sizeof(float));
/* this is for fortran, so reverse sense of slowest/fastest moving dimensions */
//for (int i=0; i<*nx; ++i) {
vmin = grid->GetValue(0,0,0);
vmax = vmin;
for (int i=0; i<*nz; ++i) {
for (int j=0; j<*ny; ++j) {
//for (int k=0; k<*nz; ++k) {
for (int k=0; k<*nx; ++k) {
//*vdata++ = grid->GetValue(i,j,k);
v = grid->GetValue(k,j,i);
if (v < 1e30 && v > -1e30) {
if (v < vmin) vmin = v;
if (v > vmax) vmax = v;
++navg;
vavg += v;
} else {
v = vmax * 1000; // just a guess
}
*vdata++ = v;
}
}
}
}
vavg = vavg/navg;
printf("min/avg/max = %f/%f/%f\n", vmin, vavg, vmax);
return 0;
}