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;
}
bool OpExtraOut::Do(OBBase* pOb, const char* OptionText, OpMap* pmap, OBConversion* pConv)
{
/*
OptionText contains an output filename with a format extension.
Make an OBConversion object with this as output destination.
Make a copy the current OBConversion and replace the output format by
an instance of ExtraFormat. This then does all the subsequent work.
*/
if(!pConv || !OptionText || *OptionText=='\0')
return true; //silent no-op. false would prevent the main output
if(pConv->IsFirstInput())
{
OBConversion* pExtraConv = new OBConversion(*pConv); //copy ensures OBConversion::Index>-1
std::ofstream* ofs;
if( (ofs = new std::ofstream(OptionText)) ) // extra parens to indicate truth value
pExtraConv->SetOutStream(ofs);
if(!ofs || !pExtraConv->SetOutFormat(OBConversion::FormatFromExt(OptionText)))
{
obErrorLog.ThrowError(__FUNCTION__, "Error setting up extra output file", obError);
return true;
}
OBConversion* pOrigConv = new OBConversion(*pConv);
//Make an instance of ExtraFormat and divert the output to it. It will delete itself.
pConv->SetOutFormat(new ExtraFormat(pOrigConv, pExtraConv));
}
return true;
}
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);
}
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);
}
// 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);
}
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);
}
// 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)
{
if (argc < 2) {
std::cout << "Usage: " << argv[0] << " <filename>" << std::endl;
return 1;
}
// Read the file
shared_ptr<OBMol> mol = GetMol(argv[1]);
// Create the OBConformerSearch object
OBConformerSearch cs;
// Setup
std::cout << "Setting up conformer searching..." << std::endl
<< " conformers: 30" << std::endl
<< " children: 5" << std::endl
<< " mutability: 5" << std::endl
<< " convergence: 25" << std::endl;
cs.Setup(*mol.get(),
30, // numConformers
5, // numChildren
5, // mutability
25); // convergence
// Perform searching
cs.Search();
// Print the rotor keys
RotorKeys keys = cs.GetRotorKeys();
for (RotorKeys::iterator key = keys.begin(); key != keys.end(); ++key) {
for (unsigned int i = 1; i < key->size(); ++i)
std::cout << key->at(i) << " ";
std::cout << std::endl;
}
// Get the conformers
cs.GetConformers(*mol.get());
std::cout << mol->NumConformers() << std::endl;
OBConversion conv;
conv.SetOutFormat("sdf");
for (unsigned int c = 0; c < mol->NumConformers(); ++c) {
mol->SetConformer(c);
conv.Write(mol.get(), &std::cerr);
}
}
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 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);
}
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;
}
///////////////////////////////////////////////////////////////////////////////
//! \brief compute rms between chemically identical molecules
int main(int argc, char **argv)
{
bool firstOnly = false;
if (argc != 3 && argc != 4)
{
cerr << "Usage: " << argv[0]
<< " [-firstonly] <reference structure(s)> <comparison structure(s)>\n";
cerr << "Computes the heavy-atom RMSD of identical compound structures.\n";
cerr << "Structures in multi-structure files are compared one-by-one unless -firstonly\n"
<< "is passed, in which case only the first structure in the reference file is used.\n";
exit(-1);
}
char *fileRef = argv[1];
char *fileTest = argv[2];
if (argc == 4)
{
//if iterate is passed as first command, try to match structures in first file to strucutres in second
if (strcmp("-firstonly", argv[1]) != 0)
{
cerr << "Usage: " << argv[0]
<< " [-firstonly] <reference structure(s)> <comparison structure(s)>\n";
exit(-1);
}
fileRef = argv[2];
fileTest = argv[3];
firstOnly = true;
}
//open mols
OBConversion refconv;
OBFormat *refFormat = refconv.FormatFromExt(fileRef);
if (!refFormat || !refconv.SetInFormat(refFormat)
|| !refconv.SetOutFormat("SMI"))
{
cerr << "Cannot read reference molecule format!" << endl;
exit(-1);
}
OBConversion testconv;
OBFormat *testFormat = testconv.FormatFromExt(fileTest);
if (!testFormat || !testconv.SetInAndOutFormats(testFormat, testFormat))
{
cerr << "Cannot read reference molecule format!" << endl;
exit(-1);
}
//read reference
ifstream ifsref;
OBMol molref;
ifsref.open(fileRef);
if (!ifsref)
{
cerr << "Cannot read fixed molecule file: " << fileRef << endl;
exit(-1);
}
//check comparison file
ifstream ifstest;
ifstest.open(fileTest);
if (!ifstest)
{
cerr << "Cannot read file: " << fileTest << endl;
exit(-1);
}
while (refconv.Read(&molref, &ifsref))
{
processMol(molref);
Matcher matcher(molref);// create the matcher
OBMol moltest;
while (testconv.Read(&moltest, &ifstest))
{
if (moltest.Empty())
break;
processMol(moltest);
double rmsd = matcher.computeRMSD(moltest);
cout << "RMSD " << moltest.GetTitle() << " " << rmsd << "\n";
if (!firstOnly)
{
break;
}
}
}
return (0);
}
int main(int argc,char *argv[])
{
OBConversion Conv(&cin, &cout); //default input and output are console
OBFormat* pInFormat = NULL;
OBFormat* pOutFormat = NULL;
bool inGzip = false;
bool outGzip = false;
vector<string> FileList, OutputFileList;
string OutputFileName;
// Parse commandline
bool gotInType = false, gotOutType = false;
bool SplitOrBatch=false;
char *oext = NULL;
char *iext = NULL;
//load plugs to fully initialize option parameters
OBPlugin::LoadAllPlugins();
//Save name of program without its path (and .exe)
string pn(argv[0]);
string::size_type pos;
#ifdef _WIN32
pos = pn.find(".exe");
if(pos!=string::npos)
argv[0][pos]='\0';
#endif
pos = pn.find_last_of("/\\");
if(pos==string::npos)
program_name=argv[0];
else
program_name=argv[0]+pos+1;
const char* p;
int arg;
for (arg = 1; arg < argc; ++arg)
{
if (argv[arg])
{
if (argv[arg][0] == '-')
{
switch (argv[arg][1])
{
case 'V':
{
cout << "Open Babel " << BABEL_VERSION << " -- "
<< __DATE__ << " -- " << __TIME__ << endl;
exit(0);
}
case 'i':
gotInType = true;
iext = argv[arg] + 2;
if(!*iext)
iext = argv[++arg]; //space left after -i: use next argument
if (strncasecmp(iext, "MIME", 4) == 0)
{
// get the MIME type from the next argument
iext = argv[++arg];
pInFormat = Conv.FormatFromMIME(iext);
}
else
{
//The ID provided by the OBFormat class is used as the identifying file extension
pInFormat = Conv.FindFormat(iext);
}
if(pInFormat==NULL)
{
cerr << program_name << ": cannot read input format!" << endl;
usage();
}
break;
case 'o':
gotOutType = true;
oext = argv[arg] + 2;
if(!*oext)
oext = argv[++arg]; //space left after -i: use next argument
if (arg >= argc)
usage(); // error in parsing command-line
if (strncasecmp(oext, "MIME", 4) == 0)
{
// get the MIME type from the next argument
oext = argv[++arg];
pOutFormat = Conv.FormatFromMIME(oext);
}
else
pOutFormat = Conv.FindFormat(oext);
if(pOutFormat==NULL)
{
cerr << program_name << ": cannot write output format!" << endl;
usage();
}
break;
case 'L': //display a list of plugin type or classes
{
const char* param=NULL;
if(argc>arg+1)
param = argv[arg+2];
// First assume first arg is a plugin type and
// param is a subtype, like babel -L ops gen3D
// or first arg is a plugin ID, like babel -L cml
OBPlugin* plugin;
if ((OBPlugin::GetPlugin("plugins", argv[arg+1]) &&
(plugin = OBPlugin::GetPlugin(argv[arg+1], param))) ||
(plugin = OBPlugin::GetPlugin(NULL, argv[arg+1])))
{
//Output details of subtype
string txt;
plugin->Display(txt, "verbose", argv[arg+1]);
cout << "One of the " << plugin->TypeID() << '\n' << txt << endl;
return 0;
}
//...otherwise assume it is a plugin type, like babel -L forcefields
//Output list of subtypes
OBPlugin::List(argv[arg+1], param);
return 0;
}
case '?':
case 'H':
if(isalnum(argv[arg][2]) || arg==argc-2)
{
if(strncasecmp(argv[arg]+2,"all",3))
{
const char* pID= (arg==argc-2) ? argv[arg+1] : argv[arg]+2;
OBFormat* pFormat = Conv.FindFormat(pID);
if(pFormat)
{
cout << pID << " " << pFormat->Description() << endl;
if(pFormat->Flags() & NOTWRITABLE)
cout << " This format is Read-only" << endl;
if(pFormat->Flags() & NOTREADABLE)
cout << " This format is Write-only" << endl;
if(strlen(pFormat->SpecificationURL()))
cout << "Specification at: " << pFormat->SpecificationURL() << endl;
}
else
cout << "Format type: " << pID << " was not recognized" <<endl;
}
else
{
OBPlugin::List("formats","verbose");
}
}
else
help();
return 0;
case '-': //long option --name text
{
//Do nothing if name is empty
//Option's text is the next arg provided it doesn't start with -
char* nam = argv[arg]+2;
if(*nam != '\0')
{
string txt;
int i;
for(i=0; i<Conv.GetOptionParams(nam, OBConversion::GENOPTIONS)
&& arg<argc-1 && argv[arg+1];++i) //removed && *argv[arg+1]!='-'
{
if(!txt.empty()) txt+=' ';
txt += argv[++arg];
}
if(*nam=='-')
{
// Is a API directive, e.g.---errorlevel
//Send to the pseudoformat "obapi" (without any leading -)
OBConversion apiConv;
OBFormat* pAPI= OBConversion::FindFormat("obapi");
if(pAPI)
{
apiConv.SetOutFormat(pAPI);
apiConv.AddOption(nam+1, OBConversion::GENOPTIONS, txt.c_str());
apiConv.Write(NULL, &std::cout);
}
}
else
// Is a long option name, e.g --addtotitle
Conv.AddOption(nam,OBConversion::GENOPTIONS,txt.c_str());
}
}
break;
case 'm': //multiple output files
SplitOrBatch=true;
break;
case 'a': //single character input option
p = argv[arg]+2;
DoOption(p,Conv,OBConversion::INOPTIONS,arg,argc,argv);
break;
case 'x': //single character output option
p = argv[arg]+2;
DoOption(p,Conv,OBConversion::OUTOPTIONS,arg,argc,argv);
break;
default: //single character general option
p = argv[arg]+1;
DoOption(p,Conv,OBConversion::GENOPTIONS,arg,argc,argv);
break;
}
}
else
{
//filenames
if(!gotOutType)
FileList.push_back(argv[arg]);
else
OutputFileName = argv[arg];
}
}
}
if(!gotOutType) //the last file is the output
{
if(FileList.empty())
{
cerr << "No output file or format spec!" << endl;
usage();
}
OutputFileName = FileList.back();
FileList.pop_back();
}
#if defined(_WIN32) && defined(USING_DYNAMIC_LIBS)
//Expand wildcards in input filenames and add to FileList
vector<string> tempFileList(FileList);
FileList.clear();
vector<string>::iterator itr;
for(itr=tempFileList.begin();itr!=tempFileList.end();++itr)
DLHandler::findFiles (FileList, *itr);
#endif
if (!gotInType)
{
if(FileList.empty())
{
cerr << "No input file or format spec!" <<endl;
usage();
}
}
if (!gotOutType)
{
pOutFormat = Conv.FormatFromExt(OutputFileName.c_str(), outGzip);
if(pOutFormat==NULL)
{
cerr << program_name << ": cannot write output format!" << endl;
usage();
}
}
if(!Conv.SetInFormat(pInFormat))
{
cerr << "Invalid input format" << endl;
usage();
}
if(!Conv.SetOutFormat(pOutFormat, outGzip))
{
cerr << "Invalid output format" << endl;
usage();
}
if(SplitOrBatch)
{
//Put * into output file name before extension (or ext.gz)
if(OutputFileName.empty())
{
OutputFileName = "*.";
if (oext != NULL)
OutputFileName += oext;
}
else
{
string::size_type pos = OutputFileName.rfind(".gz");
if(pos==string::npos)
pos = OutputFileName.rfind('.');
else
pos = OutputFileName.rfind('.',pos-1);
if(pos==string::npos)
OutputFileName += '*';
else
OutputFileName.insert(pos,"*");
}
}
int count = Conv.FullConvert(FileList, OutputFileName, OutputFileList);
Conv.ReportNumberConverted(count);
if(OutputFileList.size()>1)
{
clog << OutputFileList.size() << " files output. The first is " << OutputFileList[0] <<endl;
}
std::string messageSummary = obErrorLog.GetMessageSummary();
if (messageSummary.size())
{
clog << messageSummary << endl;
}
#ifdef _DEBUG
//CM keep window open
cout << "Press any key to finish" <<endl;
getch();
#endif
return 0;
}
bool ReactionInChIFormat::WriteMolecule(OBBase* pOb, OBConversion* pConv)
{
OBMol* pmol = dynamic_cast<OBMol*>(pOb);
if (pmol == NULL || !pmol->IsReaction())
return false;
ostream &ofs = *pConv->GetOutStream();
OBFormat* pInChIFormat = OBConversion::FindFormat("inchi");
if (!pInChIFormat)
return false;
bool isEquilibrium = pConv->IsOption("e");
OBConversion inchiconv;
inchiconv.SetOutFormat(pInChIFormat);
stringstream ss;
inchiconv.SetOutStream(&ss);
#define M_REACTANTS 0
#define M_PRODUCTS 1
#define M_AGENTS 2
OBReactionFacade facade(pmol);
std::vector<std::vector<std::string> > inchis(3);
unsigned int nonInchi[3] = { 0, 0, 0 };
bool hasNonInchi = false;
OBMol mol;
for (int part = 0; part <= 2; ++part) {
unsigned int N;
switch (part) {
case M_REACTANTS: N = facade.NumComponents(REACTANT); break;
case M_PRODUCTS: N = facade.NumComponents(PRODUCT); break;
case M_AGENTS: N = facade.NumComponents(AGENT); break;
}
for (unsigned int i = 0; i < N; ++i) {
mol.Clear();
switch (part) {
case M_REACTANTS: facade.GetComponent(&mol, REACTANT, i); break;
case M_PRODUCTS: facade.GetComponent(&mol, PRODUCT, i); break;
case M_AGENTS: facade.GetComponent(&mol, AGENT, i); break;
}
if (mol.NumAtoms() == 1 && mol.GetFirstAtom()->GetAtomicNum() == 0) {
// This represents an unknown component
nonInchi[part]++;
hasNonInchi = true;
}
else {
bool ok = inchiconv.Write(&mol);
if (!ok) {
nonInchi[part]++;
hasNonInchi = true;
}
else {
string inchi = ss.str();
if (strncmp(inchi.c_str(), "InChI=1S/", 9) != 0)
return false;
inchis[part].push_back(TrimInChI(inchi.c_str()));
}
ss.str("");
}
}
}
std::sort(inchis[M_REACTANTS].begin(), inchis[M_REACTANTS].end());
std::sort(inchis[M_PRODUCTS].begin(), inchis[M_PRODUCTS].end());
std::sort(inchis[M_AGENTS].begin(), inchis[M_AGENTS].end());
std::string reactants_string = "";
const int rsize = inchis[M_REACTANTS].size();
for (int i = 0; i < rsize; ++i) {
if (i > 0)
reactants_string += '!';
reactants_string += inchis[M_REACTANTS][i];
}
std::string products_string = "";
const int psize = inchis[M_PRODUCTS].size();
for (int i = 0; i < psize; ++i) {
if (i > 0)
products_string += '!';
products_string += inchis[M_PRODUCTS][i];
}
bool reactants_first = reactants_string <= products_string;
ofs << RINCHI_VERSION_STRING;
if (rsize > 0 || psize > 0 || !inchis[M_AGENTS].empty()) {
ofs << (reactants_first ? reactants_string : products_string);
ofs << "<>";
ofs << (reactants_first ? products_string : reactants_string);
if (!inchis[M_AGENTS].empty()) {
ofs << "<>";
for (std::vector<std::string>::const_iterator vit = inchis[M_AGENTS].begin(); vit != inchis[M_AGENTS].end(); ++vit) {
if (vit != inchis[M_AGENTS].begin())
ofs << '!';
ofs << *vit;
}
}
}
ofs << "/d";
if (isEquilibrium)
ofs << '=';
else
ofs << (reactants_first ? '+' : '-');
if (hasNonInchi) {
ofs << "/u" << (reactants_first ? nonInchi[M_REACTANTS] : nonInchi[M_PRODUCTS]) << '-'
<< (reactants_first ? nonInchi[M_PRODUCTS] : nonInchi[M_REACTANTS]) << '-'
<< nonInchi[M_AGENTS];
}
ofs << '\n';
return true;
}
// This function will call the Babel library to add
// hydrogens to the residues
void PDB::addHydrogensToPair(AminoAcid& a, AminoAcid& b, int cd1, int cd2)
{
OBMol mol;
string addedH;
istringstream tempss;
bool ligand;
if(b.atom[0]->line.find("HETATM") != string::npos)
{
ligand = true;
}
else
{
ligand = false;
}
// This section is just to suppress all of the
// warning message that aren't important to us
{
OBConversion apiConv;
OBFormat* pAPI = OBConversion::FindFormat("obapi");
if(pAPI)
{
apiConv.SetOutFormat(pAPI);
apiConv.AddOption("errorlevel", OBConversion::GENOPTIONS, "0");
apiConv.Write(NULL, &std::cout);
}
}
// Now, let's pack up the information into a string
string packedFile="";
for(unsigned int i=0; i < a.altlocs[cd1].size(); i++)
{
if( !a.altlocs[cd1][i]->skip )
{
packedFile += a.altlocs[cd1][i]->line + "\n";
}
}
int cd2_al = cd2;
if(b.residue == "ASP" || b.residue == "GLU")
{
cd2_al = cd2%(b.altlocs.size());
}
for(unsigned int i=0; i < b.altlocs[cd2_al].size(); i++)
{
if( !b.altlocs[cd2_al][i]->skip )
{
packedFile += b.altlocs[cd2_al][i]->line + "\n";
}
}
packedFile += a.makeConect(cd1);
packedFile += b.makeConect(cd2_al);
// Now, let's set up some Babel information
// First, we get the PDB format to tell
// Babel how to read the information and
// how to output it
OBFormat* pdbformat = this->conv.FindFormat("pdb");
this->conv.SetInFormat(pdbformat);
this->conv.SetOutFormat(pdbformat);
// Here is where Babel reads everything
// and adds hydrogens to the pair
// TO ADD: option to set pH
this->conv.ReadString(&mol,packedFile);
mol.AddHydrogens(false,true,PH_LEVEL);
// Let's write the newly written hydrogens to
// a string and parse it
addedH = this->conv.WriteString(&mol);
tempss.str(addedH);
// This ensures that the ligand hydrogens are labeled as
// HETATM instead of ATOM just for the sake of STAAR.
// This may be wrong, but it should be fine since we are
// stripping out that information later when we write
// the GAMESS inp files
if( ligand )
{
string line;
string f = "";
while( getline(tempss,line) )
{
if( line.find(b.residue) != string::npos )
{
line.replace(0,6,"HETATM");
}
f += line + "\n";
}
tempss.seekg(ios_base::beg);
tempss.clear();
tempss.str(f);
}
this->failure = false;
this->parsePDB(tempss,99999.99);
// This is just to ensure that all of the atoms
// are grouped together because Babel just
// appends the H to the end of the file
if( !ligand )
this->sortAtoms();
// Split the atoms up into amino acids and chains
this->populateChains(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);
}
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;
}
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 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)
{
OBForceField* pFF = OBForceField::FindForceField("Ghemical");
pFF->SetLogFile(&cout);
pFF->SetLogLevel(OBFF_LOGLVL_LOW);
OBMol mol;
mol.Clear();
char commandline[100];
vector<string> vs;
cout << endl;
cout << "openbabel " << endl;
cout << "M O L E C U L A R M E C H A N I C S" << endl;
cout << " program" << endl;
cout << " v 0.1 " << endl << endl;
while (1) {
cout << "command > ";
cin.getline(commandline, 100);
//
// commands with no parameters
//
if (EQn(commandline, "quit", 4) || cin.eof()) {
cout << "bye." << endl;
exit(0);
}
if (EQn(commandline, "help", 4) || cin.eof()) {
cout << endl;
cout << "commands: description:" << endl;
cout << "load <filename> load a molecule from filename" << endl;
cout << "save <filename> save currently loaded molecule to filename" << endl;
cout << "ff <forcefield> select the force field" << endl;
cout << "forcefields print the available forcefields" << endl;
cout << endl;
cout << "energy calculate the energy" << endl;
cout << "ebond calculate the bond stretching energy" << endl;
cout << "eangle calculate the angle bending energy" << endl;
cout << "estrbnd calculate the stretch-bending enregy" << endl;
cout << "eoop calculate the out-of-plane bending energy" << endl;
cout << "etorsion calculate the torsional energy" << endl;
cout << "evdw calculate the Van der Waals energy" << endl;
cout << "eeq calculate the electrostatic energy" << endl;
cout << endl;
cout << "sd <n> steepest descent energy minimization for n steps" << endl;
cout << "cg <n> conjugate gradients energy minimization for n steps" << endl;
cout << "" << endl;
cout << "addH add hydrogens" << endl;
cout << "delH delete hydrogens" << endl;
cout << endl;
cout << "gen generate/minimize a (random) structure" << endl;
cout << "rs rotate around all rotatable bonds" << endl;
cout << "nconf print the number of conformers" << endl;
cout << "conf <n> select conformer n" << endl;
cout << endl;
cout << "quit quit" << endl;
cout << endl;
continue;
}
// calculate the energy
if (EQn(commandline, "energy", 6)) {
if (mol.Empty()) {
cout << "no molecule loaded." << endl;
continue;
}
cout << endl << " total energy = " << pFF->Energy() << " " << pFF->GetUnit() << endl << endl;
continue;
}
if (EQn(commandline, "ebond", 5)) {
if (mol.Empty()) {
cout << "no molecule loaded." << endl;
continue;
}
cout << endl << " bond stretching energy = " << pFF->E_Bond() << " " << pFF->GetUnit() << endl << endl;
continue;
}
if (EQn(commandline, "eangle", 6)) {
if (mol.Empty()) {
cout << "no molecule loaded." << endl;
continue;
}
cout << endl << " angle bending energy = " << pFF->E_Angle() << " " << pFF->GetUnit() << endl << endl;
continue;
}
if (EQn(commandline, "estrbnd", 7)) {
if (mol.Empty()) {
cout << "no molecule loaded." << endl;
continue;
}
cout << endl << " stretch-bending energy = " << pFF->E_StrBnd() << " " << pFF->GetUnit() << endl << endl;
continue;
}
if (EQn(commandline, "eoop", 4)) {
if (mol.Empty()) {
cout << "no molecule loaded." << endl;
continue;
}
cout << endl << " out-of-plane bending energy = " << pFF->E_OOP() << " " << pFF->GetUnit() << endl << endl;
continue;
}
if (EQn(commandline, "etorsion", 8)) {
if (mol.Empty()) {
cout << "no molecule loaded." << endl;
continue;
}
cout << endl << " torsional energy = " << pFF->E_Torsion() << " " << pFF->GetUnit() << endl << endl;
continue;
}
if (EQn(commandline, "evdw", 4)) {
if (mol.Empty()) {
cout << "no molecule loaded." << endl;
continue;
}
cout << endl << " Van der Waals energy = " << pFF->E_VDW() << " " << pFF->GetUnit() << endl << endl;
continue;
}
if (EQn(commandline, "eeq", 3)) {
if (mol.Empty()) {
cout << "no molecule loaded." << endl;
continue;
}
cout << endl << " electrostatic energy = " << pFF->E_Electrostatic() << " " << pFF->GetUnit() << endl << endl;
continue;
}
if (EQn(commandline, "addH", 4)) {
int num1, num2;
num1 = mol.NumAtoms();
mol.AddHydrogens(false, true);
num2 = mol.NumAtoms();
cout << (num2 - num1) << " hydrogens added." << endl;
if (!pFF->Setup(mol)) {
cout << "error while initializing the force field for this molecule." <<endl;
continue;
}
continue;
}
if (EQn(commandline, "delH", 4)) {
int num1, num2;
num1 = mol.NumAtoms();
mol.DeleteHydrogens();
num2 = mol.NumAtoms();
cout << (num1 - num2) << " hydrogens deleted." << endl;
if (!pFF->Setup(mol)) {
cout << "error while initializing the force field for this molecule." <<endl;
continue;
}
continue;
}
if (EQn(commandline, "gen", 3)) {
//pFF->GenerateCoordinates();
pFF->UpdateCoordinates(mol);
continue;
}
if (EQn(commandline, "rs", 2)) {
pFF->SystematicRotorSearch();
pFF->UpdateCoordinates(mol);
continue;
}
if (EQn(commandline, "nconf", 5)) {
cout << endl << " number of conformers = " << mol.NumConformers() << endl << endl;
continue;
}
//
// commands with parameters
//
tokenize(vs, commandline);
// select forcefield
if (EQn(commandline, "ff", 2)) {
if (vs.size() < 2) {
cout << "no <forcefield> specified." << endl;
continue;
}
pFF = OBForceField::FindForceField(vs[1]);
if (!mol.Empty())
if (!pFF->Setup(mol))
cout << "error while initializing the force field (" << vs[1] << ") for this molecule." <<endl;
continue;
}
// load <filename>
if (EQn(commandline, "load", 4)) {
if (vs.size() < 2) {
cout << "no <filename> specified." << endl;
continue;
}
ifstream ifs;
OBConversion conv;
OBFormat *format_in = conv.FormatFromExt(vs[1].c_str());
if (!format_in || !conv.SetInFormat(format_in)) {
cout << "could not detect format." << endl;
continue;
}
ifs.open(vs[1].c_str());
if (!ifs) {
cout << "could not open '" << vs[1] << "'." <<endl;
continue;
}
mol.Clear();
if (!conv.Read(&mol, &ifs)) {
cout << "could not read a molecule from '" << vs[1] << "'." <<endl;
continue;
}
if (mol.Empty()) {
cout << "this molecule is empty." <<endl;
continue;
}
if (!pFF->Setup(mol)) {
cout << "error while initializing the force field for this molecule." <<endl;
continue;
}
cout << "molecule succesfully loaded." << endl;
cout << " " << mol.NumAtoms() << " atoms" << endl;
cout << " " << mol.NumBonds() << " bonds" << endl;
ifs.close();
continue;
}
// save <filename>
if (EQn(commandline, "save", 4)) {
if (vs.size() < 2) {
cout << "no <filename> specified." << endl;
continue;
}
ofstream ofs;
OBConversion conv;
OBFormat *format_out = conv.FormatFromExt(vs[1].c_str());
if (!format_out || !conv.SetOutFormat(format_out)) {
cout << "could not detect format." << endl;
continue;
}
ofs.open(vs[1].c_str());
if (!ofs) {
cout << "could not open '" << vs[1] << "'." <<endl;
continue;
}
if (!conv.Write(&mol, &ofs)) {
cout << "could not read a molecule from '" << vs[1] << "'." <<endl;
continue;
}
cout << "molecule succesfully saved." << endl;
cout << " " << mol.NumAtoms() << " atoms" << endl;
cout << " " << mol.NumBonds() << " bonds" << endl;
ofs.close();
continue;
}
// steepest descent
if (EQn(commandline, "sd", 2)) {
if (vs.size() < 2) {
cout << "no <n> steps specified." << endl;
continue;
}
pFF->SteepestDescent(atoi(vs[1].c_str()), OBFF_ANALYTICAL_GRADIENT);
pFF->UpdateCoordinates(mol);
continue;
}
// conjugate gradients
if (EQn(commandline, "cg", 2)) {
if (vs.size() < 2) {
cout << "no <n> steps specified." << endl;
continue;
}
pFF->ConjugateGradients(atoi(vs[1].c_str()), OBFF_ANALYTICAL_GRADIENT);
pFF->UpdateCoordinates(mol);
continue;
}
cout << "invalid command." << endl;
}
return(1);
}
int main(int argc,char *argv[])
{
OBConversion Conv(&cin, &cout); //default input and output are console
OBFormat* pInFormat = NULL;
OBFormat* pOutFormat = NULL;
vector<string> FileList, OutputFileList;
string OutputFileName;
// Parse commandline
bool gotInType = false, gotOutType = false;
bool SplitOrBatch=false;
char *oext = NULL;
char *iext = NULL;
//Save name of program without its path (and .exe)
string pn(argv[0]);
string::size_type pos;
#ifdef _WIN32
pos = pn.find(".exe");
if(pos!=string::npos)
argv[0][pos]='\0';
#endif
pos = pn.find_last_of("/\\");
if(pos==string::npos)
program_name=argv[0];
else
program_name=argv[0]+pos+1;
const char* p;
int arg;
for (arg = 1; arg < argc; ++arg)
{
if (argv[arg])
{
if (argv[arg][0] == '-')
{
char opchar[2]="?";
opchar[0]=argv[arg][1];
switch (opchar[0])
{
case 'V':
{
cout << "Open Babel " << BABEL_VERSION << " -- "
<< __DATE__ << " -- " << __TIME__ << endl;
exit(0);
}
case 'i':
//Parameter is the input format which overrides any file extensions
gotInType = true;
iext = argv[arg] + 2;
if(!*iext)
iext = argv[++arg]; //space left after -i: use next argument
if (strncasecmp(iext, "MIME", 4) == 0)
{
// get the MIME type from the next argument
iext = argv[++arg];
pInFormat = Conv.FormatFromMIME(iext);
}
else
pInFormat = Conv.FindFormat(iext);
if(pInFormat==NULL)
{
cerr << program_name << ": cannot read input format!" << endl;
usage();
}
break;
case 'o':
//Parameter is the output format which overrides any file extension
gotOutType = true;
oext = argv[arg] + 2;
if(!*oext)
oext = argv[++arg]; //space left after -i: use next argument
if (strncasecmp(oext, "MIME", 4) == 0)
{
// get the MIME type from the next argument
oext = argv[++arg];
pOutFormat = Conv.FormatFromMIME(oext);
}
else
pOutFormat = Conv.FindFormat(oext);
if(pOutFormat==NULL)
{
cerr << program_name << ": cannot write output format!" << endl;
usage();
}
break;
case 'O':
OutputFileName = argv[arg] + 2;
if(OutputFileName.size()<3)
OutputFileName = argv[++arg]; //space left after -O: use next argument
break;
case 'L': //display a list of plugin type or classes
{
const char* param=NULL;
if(argc>arg+1)
param = argv[arg+2];
// First assume first arg is a plugin type and
// param is a subtype, like babel -L ops gen3D
// or first arg is a plugin ID, like babel -L cml
OBPlugin* plugin;
if(OBPlugin::GetPlugin("plugins", argv[arg+1])
&& (plugin = OBPlugin::GetPlugin(argv[arg+1], param))
|| (plugin = OBPlugin::GetPlugin(NULL, argv[arg+1])))
{
//Output details of subtype
string txt;
plugin->Display(txt, "verbose", argv[arg+1]);
cout << "One of the " << plugin->TypeID() << '\n' << txt << endl;
return 0;
}
//...otherwise assume it is a plugin type, like babel -L forcefields
//Output list of subtypes
OBPlugin::List(argv[arg+1], param);
return 0;
}
case '?':
case 'H':
if(isalnum(argv[arg][2]) || arg==argc-2)
{
if(strncasecmp(argv[arg]+2,"all",3))
{
OBFormat* pFormat
= (arg==argc-2) ? Conv.FindFormat(argv[arg+1]) : Conv.FindFormat(argv[arg]+2);
if(pFormat)
{
cout << argv[arg]+2 << " " << pFormat->Description() << endl;
if(pFormat->Flags() & NOTWRITABLE)
cout << " This format is Read-only" << endl;
if(pFormat->Flags() & NOTREADABLE)
cout << " This format is Write-only" << endl;
if(strlen(pFormat->SpecificationURL()))
cout << "Specification at: " << pFormat->SpecificationURL() << endl;
}
else
cout << "Format type: " << argv[arg]+2 << " was not recognized" <<endl;
}
else
{
OBPlugin::List("formats","verbose");
}
}
else
help();
return 0;
case '-': //long option --name text
{
//Option's text is in the next and subsequent args, until one starts with -
char* nam = argv[arg]+2;
if(!strcasecmp(nam, "help")) //special case handled here
{
help();
return 0;
}
if(*nam != '\0') //Do nothing if name is empty
{
string txt;
while(arg<argc-1 && *argv[arg+1]!='-')
{
//use text from subsequent args
if(!txt.empty())txt += ' '; //..space separated if more than one
txt += argv[++arg];
}
// If a API directive, e.g.---errorlevel
// send to the pseudoformat "obapi" (without any leading -)
if(*nam=='-')
{
OBConversion apiConv;
OBFormat* pAPI= OBConversion::FindFormat("obapi");
if(pAPI)
{
apiConv.SetOutFormat(pAPI);
apiConv.AddOption(nam+1, OBConversion::GENOPTIONS, txt.c_str());
apiConv.Write(NULL, &std::cout);
}
}
else
// Is a normal long option name, e.g --addtotitle
Conv.AddOption(nam,OBConversion::GENOPTIONS,txt.c_str());
}
}
break;
case 'm': //multiple output files
SplitOrBatch=true;
break;
case 'a': //single character input option
p = argv[arg]+2;
DoOption(p,Conv,OBConversion::INOPTIONS,arg,argc,argv);
break;
case 'x': //single character output option
p = argv[arg]+2;
DoOption(p,Conv,OBConversion::OUTOPTIONS,arg,argc,argv);
break;
//Not essential, but allows these options to be before input filenames
//since we know they take one parameter, and are the most likely options to be misplaced
case 'f':
case 'l':
p = argv[arg] + 2;
if(!*p)
p = argv[++arg]; //space left after -f: use next argument
Conv.AddOption(opchar, OBConversion::GENOPTIONS, p);
break;
case ':':
//e.g. -:c1ccccc1. SMILES passed as a file name and handled in OBConversion
FileList.push_back(argv[arg]);
break;
default: //single character general option
p = argv[arg]+1;
DoOption(p,Conv,OBConversion::GENOPTIONS,arg,argc,argv);
break;
}
}
else //filenames
FileList.push_back(argv[arg]);
}
}
#ifdef _WIN32
//Expand wildcards in input filenames and add to FileList
vector<string> tempFileList(FileList);
FileList.clear();
vector<string>::iterator itr;
for(itr=tempFileList.begin();itr!=tempFileList.end();++itr)
{
if((*itr)[0]=='-')
FileList.push_back(*itr);
else
DLHandler::findFiles (FileList, *itr);
}
#endif
if (!gotInType)
{
if(FileList.empty())
{
cerr << "No input file or format spec or possibly a misplaced option.\n"
"Options, other than -i -o -O -m, must come after the input files.\n" <<endl;
usage();
}
}
if (!gotOutType)
{
//check there is a valid output format, but the extension will be re-interpreted in OBConversion
pOutFormat = Conv.FormatFromExt(OutputFileName.c_str());
if(OutputFileName.empty() || pOutFormat==NULL)
{
cerr << "Missing or unknown output file or format spec or possibly a misplaced option.\n"
"Options, other than -i -o -O -m, must come after the input files.\n" <<endl;
usage();
}
}
if(!Conv.SetInFormat(pInFormat))
{
cerr << "Invalid input format" << endl;
usage();
}
if(!Conv.SetOutFormat(pOutFormat))
{
cerr << "Invalid output format" << endl;
usage();
}
if(SplitOrBatch)
{
//Put * into output file name before extension (or ext.gz)
if(OutputFileName.empty())
{
OutputFileName = "*.";
OutputFileName += oext;
}
else
{
string::size_type pos = OutputFileName.rfind(".gz");
if(pos==string::npos)
pos = OutputFileName.rfind('.');
else
pos = OutputFileName.rfind('.',pos-1);
if(pos==string::npos)
OutputFileName += '*';
else
OutputFileName.insert(pos,"*");
}
}
int count = Conv.FullConvert(FileList, OutputFileName, OutputFileList);
Conv.ReportNumberConverted(count);
if(OutputFileList.size()>1)
{
clog << OutputFileList.size() << " files output. The first is " << OutputFileList[0] <<endl;
}
//std::string messageSummary = obErrorLog.GetMessageSummary();
//if (messageSummary.size())
// {
// clog << messageSummary << endl;
// }
#ifdef _DEBUG
//CM keep window open
cout << "Press any key to finish" <<endl;
getch();
#endif
return 0;
}
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);
}