src/openbabel/data.cpp

/**********************************************************************
data.cpp - Global data and resource file parsers.
 
Copyright (C) 1998-2001 by OpenEye Scientific Software, Inc.
Some portions Copyright (C) 2001-2005 by Geoffrey R. Hutchison
 
This file is part of the Open Babel project.
For more information, see <http://openbabel.sourceforge.net/>
 
This program is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation version 2 of the License.
 
This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
GNU General Public License for more details.
***********************************************************************/

#ifdef WIN32
#pragma warning (disable : 4786)
#endif

#include "babelconfig.hpp"
#include "data.hpp"
#include "mol.hpp"

// data headers
#include "element.hpp"
#include "types.hpp"
#include "isotope.hpp"
#include "resdata.hpp"


#if !HAVE_STRNCASECMP
extern "C" int strncasecmp(const char *s1, const char *s2, size_t n);
#endif

using namespace std;

namespace OpenBabel
{

OBElementTable   etab;
OBTypeTable      ttab;
OBIsotopeTable   isotab;
OBResidueData    resdat;

/** \class OBElementTable
    \brief Periodic Table of the Elements
 
    Translating element data is a common task given that many file
  formats give either element symbol or atomic number information, but
  not both. The OBElementTable class facilitates conversion between
  textual and numeric element information. An instance of the
  OBElementTable class (etab) is declared as external in data.cpp. Source
  files that include the header file mol.h automatically have an extern
  definition to etab. The following code sample demonstrates the use
  of the OBElementTable class:
\code
 cout << "The symbol for element 6 is " << etab.GetSymbol(6) << endl;
 cout << "The atomic number for Sulfur is " << etab.GetAtomicNum(16) << endl;
 cout << "The van der Waal radius for Nitrogen is " << etab.GetVdwRad(7);
\endcode
 
  Stored information in the OBElementTable includes elemental:
   - symbols
   - covalent radii
   - van der Waal radii
   - expected maximum bonding valence
   - molar mass (by IUPAC recommended atomic masses)
   - electronegativity
   - ionization potential
   - electron affinity
   - RGB colors for visualization programs
   - names (by IUPAC recommendation)
*/

OBElementTable::OBElementTable()
{
    _init = false;
    _dir = BABEL_DATADIR;
    _envvar = "BABEL_DATADIR";
    _filename = "element.txt";
    _subdir = "data";
    _dataptr = ElementData;
}

OBElementTable::~OBElementTable()
{
    vector<OBElement*>::iterator i;
    for (i = _element.begin();i != _element.end();i++)
        delete *i;
}

void OBElementTable::ParseLine(const char *buffer)
{
    int num,maxbonds;
    char symbol[5];
    char name[BUFF_SIZE];
    double Rcov,Rvdw,mass, elNeg, ionize, elAffin;
    double red, green, blue;

    if (buffer[0] != '#') // skip comment line (at the top)
    {
      sscanf(buffer,"%d %s %lf %*f %lf %d %lf %lf %lf %lf %lf %lf %lf %s",
             &num,
             symbol,
             &Rcov,
             &Rvdw,
             &maxbonds,
             &mass,
             &elNeg,
             &ionize,
             &elAffin,
             &red,
             &green,
             &blue,
             name);

      OBElement *ele = new OBElement(num,symbol,Rcov,Rvdw,maxbonds,mass,elNeg,
                                     ionize, elAffin, red, green, blue, name);
      _element.push_back(ele);
    }
}

unsigned int OBElementTable::GetNumberOfElements()
{ 
    if (!_init)
        Init();
    
    return _element.size();
}

char *OBElementTable::GetSymbol(int atomicnum)
{
    if (!_init)
        Init();

    if (atomicnum < 0 || atomicnum > static_cast<int>(_element.size()))
        return("\0");

    return(_element[atomicnum]->GetSymbol());
}

int OBElementTable::GetMaxBonds(int atomicnum)
{
    if (!_init)
        Init();

    if (atomicnum < 0 || atomicnum > static_cast<int>(_element.size()))
        return(0);

    return(_element[atomicnum]->GetMaxBonds());
}

double OBElementTable::GetElectroNeg(int atomicnum)
{
    if (!_init)
        Init();

    if (atomicnum < 0 || atomicnum > static_cast<int>(_element.size()))
        return(0.0);

    return(_element[atomicnum]->GetElectroNeg());
}

double OBElementTable::GetIonization(int atomicnum)
{
    if (!_init)
        Init();

    if (atomicnum < 0 || atomicnum > static_cast<int>(_element.size()))
        return(0.0);

    return(_element[atomicnum]->GetIonization());
}


double OBElementTable::GetElectronAffinity(int atomicnum)
{
    if (!_init)
        Init();

    if (atomicnum < 0 || atomicnum > static_cast<int>(_element.size()))
        return(0.0);

    return(_element[atomicnum]->GetElectronAffinity());
}
 
vector<double> OBElementTable::GetRGB(int atomicnum)
{
    if (!_init)
        Init();

    vector <double> colors;
    colors.reserve(3);
    
    if (atomicnum < 0 || atomicnum > static_cast<int>(_element.size()))
      {
        colors.push_back(0.0f);
        colors.push_back(0.0f);
        colors.push_back(0.0f);
        return(colors);
      }

    colors.push_back(_element[atomicnum]->GetRed());
    colors.push_back(_element[atomicnum]->GetGreen());
    colors.push_back(_element[atomicnum]->GetBlue());

    return (colors);
}
 
string OBElementTable::GetName(int atomicnum)
{
    if (!_init)
        Init();

    if (atomicnum < 0 || atomicnum > static_cast<int>(_element.size()))
        return("Unknown");

    return(_element[atomicnum]->GetName());
}

double OBElementTable::GetVdwRad(int atomicnum)
{
    if (!_init)
        Init();

    if (atomicnum < 0 || atomicnum > static_cast<int>(_element.size()))
        return(0.0);

    return(_element[atomicnum]->GetVdwRad());
}

double OBElementTable::CorrectedBondRad(int atomicnum, int hyb)
{
    double rad;
    if (!_init)
        Init();

    if (atomicnum < 0 || atomicnum > static_cast<int>(_element.size()))
        return(1.0);

    rad = _element[atomicnum]->GetCovalentRad();

    if (hyb == 2)
        rad *= 0.95;
    else if (hyb == 1)
        rad *= 0.90;

    return(rad);
}

double OBElementTable::CorrectedVdwRad(int atomicnum, int hyb)
{
    double rad;
    if (!_init)
        Init();

    if (atomicnum < 0 || atomicnum > static_cast<int>(_element.size()))
        return(1.95);

    rad = _element[atomicnum]->GetVdwRad();

    if (hyb == 2)
        rad *= 0.95;
    else if (hyb == 1)
        rad *= 0.90;

    return(rad);
}

double OBElementTable::GetCovalentRad(int atomicnum)
{
    if (!_init)
        Init();

    if (atomicnum < 0 || atomicnum > static_cast<int>(_element.size()))
        return(0.0);

    return(_element[atomicnum]->GetCovalentRad());
}

double OBElementTable::GetMass(int atomicnum)
{
    if (!_init)
        Init();

    if (atomicnum < 0 || atomicnum > static_cast<int>(_element.size()))
        return(0.0);

    return(_element[atomicnum]->GetMass());
}

int OBElementTable::GetAtomicNum(const char *sym)
{
  int temp;
  return GetAtomicNum(sym, temp);
}

int OBElementTable::GetAtomicNum(const char *sym, int &iso)
{
    if (!_init)
        Init();

    vector<OBElement*>::iterator i;
    for (i = _element.begin();i != _element.end();i++)
        if (!strncasecmp(sym,(*i)->GetSymbol(),2))
            return((*i)->GetAtomicNum());
    if (strcasecmp(sym, "D") == 0)
    {
        iso = 2;
        return(1);
    }
    else if (strcasecmp(sym, "T") == 0)
    {
        iso = 3;
        return(1);
    }
                else
                        iso = 0;
    return(0);
}

/** \class OBIsotopeTable
    \brief Table of atomic isotope masses
 
*/

OBIsotopeTable::OBIsotopeTable()
{
    _init = false;
    _dir = BABEL_DATADIR;
    _envvar = "BABEL_DATADIR";
    _filename = "isotope.txt";
    _subdir = "data";
    _dataptr = IsotopeData;
}

void OBIsotopeTable::ParseLine(const char *buffer)
{
    unsigned int atomicNum;
    unsigned int i;
    vector<string> vs;

    pair <unsigned int, double> entry;
    vector <pair <unsigned int, double> > row;

    if (buffer[0] != '#') // skip comment line (at the top)
    {
        tokenize(vs,buffer);
        if (vs.size() > 3) // atomic number, 0, most abundant mass (...)
        {
            atomicNum = atoi(vs[0].c_str());
            for (i = 1; i < vs.size() - 1; i += 2) // make sure i+1 still exists
            {
                entry.first = atoi(vs[i].c_str()); // isotope
                entry.second = atof(vs[i + 1].c_str()); // exact mass
                row.push_back(entry);
            }
            _isotopes.push_back(row);
        }
        else
          obErrorLog.ThrowError(__FUNCTION__, " Could not parse line in isotope table isotope.txt", obInfo);
    }
}

double  OBIsotopeTable::GetExactMass(const unsigned int ele,
                                    const unsigned int isotope)
{
    if (!_init)
        Init();

    if (ele > _isotopes.size())
        return 0.0;

    unsigned int iso;
    for (iso = 0; iso < _isotopes[ele].size(); iso++)
        if (isotope == _isotopes[ele][iso].first)
            return _isotopes[ele][iso].second;

    return 0.0;
}

/** \class OBTypeTable
    \brief Atom Type Translation Table
 
Molecular file formats frequently store information about atoms in an
atom type field. Some formats store only the element for each atom,
while others include hybridization and local environments, such as the
Sybyl mol2 atom type field. The OBTypeTable class acts as a translation
table to convert atom types between a number of different molecular
file formats. The constructor for OBTypeTable automatically reads the
text file types.txt. Just as OBElementTable, an instance of
OBTypeTable (ttab) is declared external in data.cpp and is referenced as
extern OBTypeTable ttab in mol.h.  The following code demonstrates how
to use the OBTypeTable class to translate the internal representation
of atom types in an OBMol Internal to Sybyl Mol2 atom types.
 
\code
ttab.SetFromType("INT");
ttab.SetToType("SYB");
OBAtom *atom;
vector<OBAtom*>::iterator i;
string src,dst;
for (atom = mol.BeginAtom(i);atom;atom = mol.EndAtom(i))
{
src = atom->GetType();
ttab.Translate(dst,src);
cout << "atom number " << atom->GetIdx() << "has mol2 type " << dst << endl;
}
\endcode
 
Current atom types include (defined in the top line of the data file types.txt):
- INT (Open Babel internal codes)
- ATN (atomic numbers)
- HYB (hybridization)
- MMD
- MM2 (MM2 force field)
- XYZ (element symbols from XYZ file format)
- ALC (Alchemy file)
- HAD
- MCML
- C3D (Chem3D)
- SYB (Sybyl mol2)
- MOL
- MAP
- DRE
- XED (XED format)
- DOK (Dock)
- M3D
*/

OBTypeTable::OBTypeTable()
{
    _init = false;
    _dir = BABEL_DATADIR;
    _envvar = "BABEL_DATADIR";
    _filename = "types.txt";
    _subdir = "data";
    _dataptr = TypesData;
    _linecount = 0;
    _from = _to = -1;
}

void OBTypeTable::ParseLine(const char *buffer)
{
    if (buffer[0] == '#')
        return; // just a comment line

    if (_linecount == 0)
        sscanf(buffer,"%d%d",&_nrows,&_ncols);
    else if (_linecount == 1)
        tokenize(_colnames,buffer);
    else
    {
        vector<string> vc;
        tokenize(vc,buffer);
        if (vc.size() == (unsigned)_ncols)
            _table.push_back(vc);
        else
          {
            stringstream errorMsg;
            errorMsg << " Could not parse line in type translation table types.txt -- incorect number of columns";
            errorMsg << " found " << vc.size() << " expected " << _ncols << ".";
            obErrorLog.ThrowError(__FUNCTION__, errorMsg.str(), obInfo);
          }
    }
    _linecount++;
}

bool OBTypeTable::SetFromType(const char* from)
{
    if (!_init)
        Init();

    string tmp = from;

    unsigned int i;
    for (i = 0;i < _colnames.size();i++)
        if (tmp == _colnames[i])
        {
            _from = i;
            return(true);
        }

    obErrorLog.ThrowError(__FUNCTION__, "Requested type column not found", obInfo);

    return(false);
}

bool OBTypeTable::SetToType(const char* to)
{
    if (!_init)
        Init();

    string tmp = to;

    unsigned int i;
    for (i = 0;i < _colnames.size();i++)
        if (tmp == _colnames[i])
        {
            _to = i;
            return(true);
        }

    obErrorLog.ThrowError(__FUNCTION__, "Requested type column not found", obInfo);

    return(false);
}

bool OBTypeTable::Translate(char *to, const char *from)
{
    if (!_init)
        Init();

    bool rval;
    string sto,sfrom;
    sfrom = from;
    rval = Translate(sto,sfrom);
    strcpy(to,(char*)sto.c_str());

    return(rval);
}

bool OBTypeTable::Translate(string &to, const string &from)
{
    if (!_init)
        Init();

    if (from == "")
        return(false);

    if (_from >= 0 && _to >= 0 &&
        _from < _table.size() && _to < _table.size())
      {
        vector<vector<string> >::iterator i;
        for (i = _table.begin();i != _table.end();i++)
          if ((signed)(*i).size() > _from &&  (*i)[_from] == from)
            {
              to = (*i)[_to];
              return(true);
            }
      }

    // Throw an error, copy the string and return false
    obErrorLog.ThrowError(__FUNCTION__, "Cannot perform atom type translation: table cannot find requested types.", obWarning);
    to = from;
    return(false);
}

std::string OBTypeTable::GetFromType()
{
    if (!_init)
        Init();

    if (_from > 0 && _from < _table.size())
      return( _colnames[_from] );
    else
      return( _colnames[0] );
}

std::string OBTypeTable::GetToType()
{
    if (!_init)
        Init();

    if (_to > 0 && _to < _table.size())
      return( _colnames[_to] );
    else
      return( _colnames[0] );
}

void Toupper(string &s)
{
    unsigned int i;
    for (i = 0;i < s.size();i++)
        s[i] = toupper(s[i]);
}

void Tolower(string &s)
{
    unsigned int i;
    for (i = 0;i < s.size();i++)
        s[i] = tolower(s[i]);
}

///////////////////////////////////////////////////////////////////////
OBResidueData::OBResidueData()
{
    _init = false;
    _dir = BABEL_DATADIR;
    _envvar = "BABEL_DATADIR";
    _filename = "resdata.txt";
    _subdir = "data";
    _dataptr = ResidueData;
}

bool OBResidueData::AssignBonds(OBMol &mol,OBBitVec &bv)
{
    OBAtom *a1,*a2;
    OBResidue *r1,*r2;
    vector<OBNodeBase*>::iterator i,j;
    vector3 v;

    int bo;
    unsigned int skipres=0;
    string rname = "";
    //assign residue bonds
    for (a1 = mol.BeginAtom(i);a1;a1 = mol.NextAtom(i))
    {
        r1 = a1->GetResidue();
        if (skipres && r1->GetNum() == skipres)
            continue;

        if (r1->GetName() != rname)
        {
            skipres = SetResName(r1->GetName()) ? 0 : r1->GetNum();
            rname = r1->GetName();
        }
        //assign bonds for each atom
        for (j=i,a2 = mol.NextAtom(j);a2;a2 = mol.NextAtom(j))
        {
            r2 = a2->GetResidue();
            if (r1->GetNum() != r2->GetNum())
                break;
            if (r1->GetName() != r2->GetName())
                break;

            if ((bo = LookupBO(r1->GetAtomID(a1),r2->GetAtomID(a2))))
            {
                v = a1->GetVector() - a2->GetVector();
                if (v.length_2() < 3.5) //check by distance
                    mol.AddBond(a1->GetIdx(),a2->GetIdx(),bo);
            }
        }
    }

    int hyb;
    string type;

    //types and hybridization
    for (a1 = mol.BeginAtom(i);a1;a1 = mol.NextAtom(i))
    {
        if (a1->IsOxygen() && !a1->GetValence())
        {
            a1->SetType("O3");
            continue;
        }

        if (a1->IsHydrogen())
        {
            a1->SetType("H");
            continue;
        }

        r1 = a1->GetResidue();
        if (skipres && r1->GetNum() == skipres)
            continue;

        if (r1->GetName() != rname)
        {
            skipres = SetResName(r1->GetName()) ? 0 : r1->GetNum();
            rname = r1->GetName();
        }

        //***valence rule for O-
        if (a1->IsOxygen() && a1->GetValence() == 1)
        {
            OBBond *bond;
            bond = (OBBond*)*(a1->BeginBonds());
            if (bond->GetBO() == 2)
            {
                a1->SetType("O2");
                a1->SetHyb(2);
            }
            if (bond->GetBO() == 1)
            {
                a1->SetType("O-");
                a1->SetHyb(3);
                a1->SetFormalCharge(-1);
            }
        }
        else
            if (LookupType(r1->GetAtomID(a1),type,hyb))
            {
                a1->SetType(type);
                a1->SetHyb(hyb);
            }
            else // try to figure it out by bond order ???
            {}
    }

    return(true);
}

void OBResidueData::ParseLine(const char *buffer)
{
    int bo;
    string s;
    vector<string> vs;

    if (buffer[0] == '#')
        return;

    tokenize(vs,buffer);
    if (!vs.empty())
    {
        if (vs[0] == "BOND")
        {
            s = (vs[1] < vs[2]) ? vs[1] + " " + vs[2] :
                vs[2] + " " + vs[1];
            bo = atoi(vs[3].c_str());
            _vtmp.push_back(pair<string,int> (s,bo));
        }

        if (vs[0] == "ATOM" && vs.size() == 4)
        {
            _vatmtmp.push_back(vs[1]);
            _vatmtmp.push_back(vs[2]);
            _vatmtmp.push_back(vs[3]);
        }

        if (vs[0] == "RES")
            _resname.push_back(vs[1]);

        if (vs[0]== "END")
        {
            _resatoms.push_back(_vatmtmp);
            _resbonds.push_back(_vtmp);
            _vtmp.clear();
            _vatmtmp.clear();
        }
    }
}

bool OBResidueData::SetResName(const string &s)
{
    unsigned int i;
    for (i = 0;i < _resname.size();i++)
        if (_resname[i] == s)
        {
            _resnum = i;
            return(true);
        }

    _resnum = -1;
    return(false);
}

int OBResidueData::LookupBO(const string &s)
{
    if (_resnum == -1)
        return(0);

    unsigned int i;
    for (i = 0;i < _resbonds[_resnum].size();i++)
        if (_resbonds[_resnum][i].first == s)
            return(_resbonds[_resnum][i].second);

    return(0);
}

int OBResidueData::LookupBO(const string &s1, const string &s2)
{
    if (_resnum == -1)
        return(0);
    string s;

    s = (s1 < s2) ? s1 + " " + s2 : s2 + " " + s1;

    unsigned int i;
    for (i = 0;i < _resbonds[_resnum].size();i++)
        if (_resbonds[_resnum][i].first == s)
            return(_resbonds[_resnum][i].second);

    return(0);
}

bool OBResidueData::LookupType(const string &atmid,string &type,int &hyb)
{
    if (_resnum == -1)
        return(false);

    string s;
    vector<string>::iterator i;

    for (i = _resatoms[_resnum].begin();i != _resatoms[_resnum].end();i+=3)
        if (atmid == *i)
        {
            i++;
            type = *i;
            i++;
            hyb = atoi((*i).c_str());
            return(true);
        }

    return(false);
}

void OBGlobalDataBase::Init()
{
    if (_init)
        return;
    _init = true;

    char buffer[BUFF_SIZE],subbuffer[BUFF_SIZE];
    ifstream ifs1, ifs2, ifs3, ifs4, *ifsP;
    // First, look for an environment variable
    if (getenv(_envvar.c_str()) != NULL)
    {
        strcpy(buffer,getenv(_envvar.c_str()));
        strcat(buffer,FILE_SEP_CHAR);

        if (!_subdir.empty())
        {
            strcpy(subbuffer,buffer);
            strcat(subbuffer,_subdir.c_str());
            strcat(subbuffer,FILE_SEP_CHAR);
        }

        strcat(buffer,(char*)_filename.c_str());
        strcat(subbuffer,(char*)_filename.c_str());

        ifs1.open(subbuffer);
        ifsP= &ifs1;
        if (!(*ifsP))
        {
            ifs2.open(buffer);
            ifsP = &ifs2;
        }
    }
    // Then, check the configured data directory
    else // if (!(*ifsP))
    {
        strcpy(buffer,_dir.c_str());
        strcat(buffer,FILE_SEP_CHAR);

        strcpy(subbuffer,buffer);
        strcat(subbuffer,BABEL_VERSION);
        strcat(subbuffer,FILE_SEP_CHAR);
        strcat(subbuffer,(char*)_filename.c_str());

        strcat(buffer,(char*)_filename.c_str());

        ifs3.open(subbuffer);
        ifsP= &ifs3;
        if (!(*ifsP))
        {
            ifs4.open(buffer);
            ifsP = &ifs4;
        }
    }

    if ((*ifsP))
    {
        while(ifsP->getline(buffer,BUFF_SIZE))
            ParseLine(buffer);
    }

    else
        // If all else fails, use the compiled in values
        if (_dataptr)
        {
            const char *p1,*p2;
            for (p1 = p2 = _dataptr;*p2 != '\0';p2++)
                if (*p2 == '\n')
                {
                    strncpy(buffer, p1, (p2 - p1));
                    buffer[(p2 - p1)] = '\0';
                    ParseLine(buffer);
                    p1 = ++p2;
                }
        }
        else
        {
            string s = "Unable to open data file '";
            s += _filename;
            s += "'";
            obErrorLog.ThrowError(__FUNCTION__, s, obWarning);
        }

    if (ifs1)
        ifs1.close();
    if (ifs2)
        ifs2.close();
    if (ifs3)
        ifs3.close();
    if (ifs4)
        ifs4.close();

    if (GetSize() == 0)
      {
            string s = "Cannot initialize database '";
            s += _filename;
            s += "' which may cause further errors.";
      obErrorLog.ThrowError(__FUNCTION__, "Cannot initialize database", obWarning);
      }
      
}

} // end namespace OpenBabel

//! \file data.cpp
//! \brief Global data and resource file parsers.
Revision:	2440
Committed:	Wed Nov 16 19:42:11 2005 UTC (18 years, 8 months ago) by tim
File size:	21365 byte(s)
Log Message:	adding openbabel
#	User	Rev	Content
1	tim	2440	/**********************************************************************
2			data.cpp - Global data and resource file parsers.
3
4			Copyright (C) 1998-2001 by OpenEye Scientific Software, Inc.
5			Some portions Copyright (C) 2001-2005 by Geoffrey R. Hutchison
6
7			This file is part of the Open Babel project.
8			For more information, see <http://openbabel.sourceforge.net/>
9
10			This program is free software; you can redistribute it and/or modify
11			it under the terms of the GNU General Public License as published by
12			the Free Software Foundation version 2 of the License.
13
14			This program is distributed in the hope that it will be useful,
15			but WITHOUT ANY WARRANTY; without even the implied warranty of
16			MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
17			GNU General Public License for more details.
18			***********************************************************************/
19
20			#ifdef WIN32
21			#pragma warning (disable : 4786)
22			#endif
23
24			#include "babelconfig.hpp"
25			#include "data.hpp"
26			#include "mol.hpp"
27
28			// data headers
29			#include "element.hpp"
30			#include "types.hpp"
31			#include "isotope.hpp"
32			#include "resdata.hpp"
33
34
35			#if !HAVE_STRNCASECMP
36			extern "C" int strncasecmp(const char s1, const char s2, size_t n);
37			#endif
38
39			using namespace std;
40
41			namespace OpenBabel
42			{
43
44			OBElementTable etab;
45			OBTypeTable ttab;
46			OBIsotopeTable isotab;
47			OBResidueData resdat;
48
49			/** \class OBElementTable
50			\brief Periodic Table of the Elements
51
52			Translating element data is a common task given that many file
53			formats give either element symbol or atomic number information, but
54			not both. The OBElementTable class facilitates conversion between
55			textual and numeric element information. An instance of the
56			OBElementTable class (etab) is declared as external in data.cpp. Source
57			files that include the header file mol.h automatically have an extern
58			definition to etab. The following code sample demonstrates the use
59			of the OBElementTable class:
60			\code
61			cout << "The symbol for element 6 is " << etab.GetSymbol(6) << endl;
62			cout << "The atomic number for Sulfur is " << etab.GetAtomicNum(16) << endl;
63			cout << "The van der Waal radius for Nitrogen is " << etab.GetVdwRad(7);
64			\endcode
65
66			Stored information in the OBElementTable includes elemental:
67			- symbols
68			- covalent radii
69			- van der Waal radii
70			- expected maximum bonding valence
71			- molar mass (by IUPAC recommended atomic masses)
72			- electronegativity
73			- ionization potential
74			- electron affinity
75			- RGB colors for visualization programs
76			- names (by IUPAC recommendation)
77			*/
78
79			OBElementTable::OBElementTable()
80			{
81			_init = false;
82			_dir = BABEL_DATADIR;
83			_envvar = "BABEL_DATADIR";
84			_filename = "element.txt";
85			_subdir = "data";
86			_dataptr = ElementData;
87			}
88
89			OBElementTable::~OBElementTable()
90			{
91			vector<OBElement*>::iterator i;
92			for (i = _element.begin();i != _element.end();i++)
93			delete *i;
94			}
95
96			void OBElementTable::ParseLine(const char *buffer)
97			{
98			int num,maxbonds;
99			char symbol[5];
100			char name[BUFF_SIZE];
101			double Rcov,Rvdw,mass, elNeg, ionize, elAffin;
102			double red, green, blue;
103
104			if (buffer[0] != '#') // skip comment line (at the top)
105			{
106			sscanf(buffer,"%d %s %lf %*f %lf %d %lf %lf %lf %lf %lf %lf %lf %s",
107			&num,
108			symbol,
109			&Rcov,
110			&Rvdw,
111			&maxbonds,
112			&mass,
113			&elNeg,
114			&ionize,
115			&elAffin,
116			&red,
117			&green,
118			&blue,
119			name);
120
121			OBElement *ele = new OBElement(num,symbol,Rcov,Rvdw,maxbonds,mass,elNeg,
122			ionize, elAffin, red, green, blue, name);
123			_element.push_back(ele);
124			}
125			}
126
127			unsigned int OBElementTable::GetNumberOfElements()
128			{
129			if (!_init)
130			Init();
131
132			return _element.size();
133			}
134
135			char *OBElementTable::GetSymbol(int atomicnum)
136			{
137			if (!_init)
138			Init();
139
140			if (atomicnum < 0 \|\| atomicnum > static_cast<int>(_element.size()))
141			return("\0");
142
143			return(_element[atomicnum]->GetSymbol());
144			}
145
146			int OBElementTable::GetMaxBonds(int atomicnum)
147			{
148			if (!_init)
149			Init();
150
151			if (atomicnum < 0 \|\| atomicnum > static_cast<int>(_element.size()))
152			return(0);
153
154			return(_element[atomicnum]->GetMaxBonds());
155			}
156
157			double OBElementTable::GetElectroNeg(int atomicnum)
158			{
159			if (!_init)
160			Init();
161
162			if (atomicnum < 0 \|\| atomicnum > static_cast<int>(_element.size()))
163			return(0.0);
164
165			return(_element[atomicnum]->GetElectroNeg());
166			}
167
168			double OBElementTable::GetIonization(int atomicnum)
169			{
170			if (!_init)
171			Init();
172
173			if (atomicnum < 0 \|\| atomicnum > static_cast<int>(_element.size()))
174			return(0.0);
175
176			return(_element[atomicnum]->GetIonization());
177			}
178
179
180			double OBElementTable::GetElectronAffinity(int atomicnum)
181			{
182			if (!_init)
183			Init();
184
185			if (atomicnum < 0 \|\| atomicnum > static_cast<int>(_element.size()))
186			return(0.0);
187
188			return(_element[atomicnum]->GetElectronAffinity());
189			}
190
191			vector<double> OBElementTable::GetRGB(int atomicnum)
192			{
193			if (!_init)
194			Init();
195
196			vector <double> colors;
197			colors.reserve(3);
198
199			if (atomicnum < 0 \|\| atomicnum > static_cast<int>(_element.size()))
200			{
201			colors.push_back(0.0f);
202			colors.push_back(0.0f);
203			colors.push_back(0.0f);
204			return(colors);
205			}
206
207			colors.push_back(_element[atomicnum]->GetRed());
208			colors.push_back(_element[atomicnum]->GetGreen());
209			colors.push_back(_element[atomicnum]->GetBlue());
210
211			return (colors);
212			}
213
214			string OBElementTable::GetName(int atomicnum)
215			{
216			if (!_init)
217			Init();
218
219			if (atomicnum < 0 \|\| atomicnum > static_cast<int>(_element.size()))
220			return("Unknown");
221
222			return(_element[atomicnum]->GetName());
223			}
224
225			double OBElementTable::GetVdwRad(int atomicnum)
226			{
227			if (!_init)
228			Init();
229
230			if (atomicnum < 0 \|\| atomicnum > static_cast<int>(_element.size()))
231			return(0.0);
232
233			return(_element[atomicnum]->GetVdwRad());
234			}
235
236			double OBElementTable::CorrectedBondRad(int atomicnum, int hyb)
237			{
238			double rad;
239			if (!_init)
240			Init();
241
242			if (atomicnum < 0 \|\| atomicnum > static_cast<int>(_element.size()))
243			return(1.0);
244
245			rad = _element[atomicnum]->GetCovalentRad();
246
247			if (hyb == 2)
248			rad *= 0.95;
249			else if (hyb == 1)
250			rad *= 0.90;
251
252			return(rad);
253			}
254
255			double OBElementTable::CorrectedVdwRad(int atomicnum, int hyb)
256			{
257			double rad;
258			if (!_init)
259			Init();
260
261			if (atomicnum < 0 \|\| atomicnum > static_cast<int>(_element.size()))
262			return(1.95);
263
264			rad = _element[atomicnum]->GetVdwRad();
265
266			if (hyb == 2)
267			rad *= 0.95;
268			else if (hyb == 1)
269			rad *= 0.90;
270
271			return(rad);
272			}
273
274			double OBElementTable::GetCovalentRad(int atomicnum)
275			{
276			if (!_init)
277			Init();
278
279			if (atomicnum < 0 \|\| atomicnum > static_cast<int>(_element.size()))
280			return(0.0);
281
282			return(_element[atomicnum]->GetCovalentRad());
283			}
284
285			double OBElementTable::GetMass(int atomicnum)
286			{
287			if (!_init)
288			Init();
289
290			if (atomicnum < 0 \|\| atomicnum > static_cast<int>(_element.size()))
291			return(0.0);
292
293			return(_element[atomicnum]->GetMass());
294			}
295
296			int OBElementTable::GetAtomicNum(const char *sym)
297			{
298			int temp;
299			return GetAtomicNum(sym, temp);
300			}
301
302			int OBElementTable::GetAtomicNum(const char *sym, int &iso)
303			{
304			if (!_init)
305			Init();
306
307			vector<OBElement*>::iterator i;
308			for (i = _element.begin();i != _element.end();i++)
309			if (!strncasecmp(sym,(*i)->GetSymbol(),2))
310			return((*i)->GetAtomicNum());
311			if (strcasecmp(sym, "D") == 0)
312			{
313			iso = 2;
314			return(1);
315			}
316			else if (strcasecmp(sym, "T") == 0)
317			{
318			iso = 3;
319			return(1);
320			}
321			else
322			iso = 0;
323			return(0);
324			}
325
326			/** \class OBIsotopeTable
327			\brief Table of atomic isotope masses
328
329			*/
330
331			OBIsotopeTable::OBIsotopeTable()
332			{
333			_init = false;
334			_dir = BABEL_DATADIR;
335			_envvar = "BABEL_DATADIR";
336			_filename = "isotope.txt";
337			_subdir = "data";
338			_dataptr = IsotopeData;
339			}
340
341			void OBIsotopeTable::ParseLine(const char *buffer)
342			{
343			unsigned int atomicNum;
344			unsigned int i;
345			vector<string> vs;
346
347			pair <unsigned int, double> entry;
348			vector <pair <unsigned int, double> > row;
349
350			if (buffer[0] != '#') // skip comment line (at the top)
351			{
352			tokenize(vs,buffer);
353			if (vs.size() > 3) // atomic number, 0, most abundant mass (...)
354			{
355			atomicNum = atoi(vs[0].c_str());
356			for (i = 1; i < vs.size() - 1; i += 2) // make sure i+1 still exists
357			{
358			entry.first = atoi(vs[i].c_str()); // isotope
359			entry.second = atof(vs[i + 1].c_str()); // exact mass
360			row.push_back(entry);
361			}
362			_isotopes.push_back(row);
363			}
364			else
365			obErrorLog.ThrowError(__FUNCTION__, " Could not parse line in isotope table isotope.txt", obInfo);
366			}
367			}
368
369			double OBIsotopeTable::GetExactMass(const unsigned int ele,
370			const unsigned int isotope)
371			{
372			if (!_init)
373			Init();
374
375			if (ele > _isotopes.size())
376			return 0.0;
377
378			unsigned int iso;
379			for (iso = 0; iso < _isotopes[ele].size(); iso++)
380			if (isotope == _isotopes[ele][iso].first)
381			return _isotopes[ele][iso].second;
382
383			return 0.0;
384			}
385
386			/** \class OBTypeTable
387			\brief Atom Type Translation Table
388
389			Molecular file formats frequently store information about atoms in an
390			atom type field. Some formats store only the element for each atom,
391			while others include hybridization and local environments, such as the
392			Sybyl mol2 atom type field. The OBTypeTable class acts as a translation
393			table to convert atom types between a number of different molecular
394			file formats. The constructor for OBTypeTable automatically reads the
395			text file types.txt. Just as OBElementTable, an instance of
396			OBTypeTable (ttab) is declared external in data.cpp and is referenced as
397			extern OBTypeTable ttab in mol.h. The following code demonstrates how
398			to use the OBTypeTable class to translate the internal representation
399			of atom types in an OBMol Internal to Sybyl Mol2 atom types.
400
401			\code
402			ttab.SetFromType("INT");
403			ttab.SetToType("SYB");
404			OBAtom *atom;
405			vector<OBAtom*>::iterator i;
406			string src,dst;
407			for (atom = mol.BeginAtom(i);atom;atom = mol.EndAtom(i))
408			{
409			src = atom->GetType();
410			ttab.Translate(dst,src);
411			cout << "atom number " << atom->GetIdx() << "has mol2 type " << dst << endl;
412			}
413			\endcode
414
415			Current atom types include (defined in the top line of the data file types.txt):
416			- INT (Open Babel internal codes)
417			- ATN (atomic numbers)
418			- HYB (hybridization)
419			- MMD
420			- MM2 (MM2 force field)
421			- XYZ (element symbols from XYZ file format)
422			- ALC (Alchemy file)
423			- HAD
424			- MCML
425			- C3D (Chem3D)
426			- SYB (Sybyl mol2)
427			- MOL
428			- MAP
429			- DRE
430			- XED (XED format)
431			- DOK (Dock)
432			- M3D
433			*/
434
435			OBTypeTable::OBTypeTable()
436			{
437			_init = false;
438			_dir = BABEL_DATADIR;
439			_envvar = "BABEL_DATADIR";
440			_filename = "types.txt";
441			_subdir = "data";
442			_dataptr = TypesData;
443			_linecount = 0;
444			_from = _to = -1;
445			}
446
447			void OBTypeTable::ParseLine(const char *buffer)
448			{
449			if (buffer[0] == '#')
450			return; // just a comment line
451
452			if (_linecount == 0)
453			sscanf(buffer,"%d%d",&_nrows,&_ncols);
454			else if (_linecount == 1)
455			tokenize(_colnames,buffer);
456			else
457			{
458			vector<string> vc;
459			tokenize(vc,buffer);
460			if (vc.size() == (unsigned)_ncols)
461			_table.push_back(vc);
462			else
463			{
464			stringstream errorMsg;
465			errorMsg << " Could not parse line in type translation table types.txt -- incorect number of columns";
466			errorMsg << " found " << vc.size() << " expected " << _ncols << ".";
467			obErrorLog.ThrowError(__FUNCTION__, errorMsg.str(), obInfo);
468			}
469			}
470			_linecount++;
471			}
472
473			bool OBTypeTable::SetFromType(const char* from)
474			{
475			if (!_init)
476			Init();
477
478			string tmp = from;
479
480			unsigned int i;
481			for (i = 0;i < _colnames.size();i++)
482			if (tmp == _colnames[i])
483			{
484			_from = i;
485			return(true);
486			}
487
488			obErrorLog.ThrowError(__FUNCTION__, "Requested type column not found", obInfo);
489
490			return(false);
491			}
492
493			bool OBTypeTable::SetToType(const char* to)
494			{
495			if (!_init)
496			Init();
497
498			string tmp = to;
499
500			unsigned int i;
501			for (i = 0;i < _colnames.size();i++)
502			if (tmp == _colnames[i])
503			{
504			_to = i;
505			return(true);
506			}
507
508			obErrorLog.ThrowError(__FUNCTION__, "Requested type column not found", obInfo);
509
510			return(false);
511			}
512
513			bool OBTypeTable::Translate(char to, const char from)
514			{
515			if (!_init)
516			Init();
517
518			bool rval;
519			string sto,sfrom;
520			sfrom = from;
521			rval = Translate(sto,sfrom);
522			strcpy(to,(char*)sto.c_str());
523
524			return(rval);
525			}
526
527			bool OBTypeTable::Translate(string &to, const string &from)
528			{
529			if (!_init)
530			Init();
531
532			if (from == "")
533			return(false);
534
535			if (_from >= 0 && _to >= 0 &&
536			_from < _table.size() && _to < _table.size())
537			{
538			vector<vector<string> >::iterator i;
539			for (i = _table.begin();i != _table.end();i++)
540			if ((signed)(i).size() > _from && (i)[_from] == from)
541			{
542			to = (*i)[_to];
543			return(true);
544			}
545			}
546
547			// Throw an error, copy the string and return false
548			obErrorLog.ThrowError(__FUNCTION__, "Cannot perform atom type translation: table cannot find requested types.", obWarning);
549			to = from;
550			return(false);
551			}
552
553			std::string OBTypeTable::GetFromType()
554			{
555			if (!_init)
556			Init();
557
558			if (_from > 0 && _from < _table.size())
559			return( _colnames[_from] );
560			else
561			return( _colnames[0] );
562			}
563
564			std::string OBTypeTable::GetToType()
565			{
566			if (!_init)
567			Init();
568
569			if (_to > 0 && _to < _table.size())
570			return( _colnames[_to] );
571			else
572			return( _colnames[0] );
573			}
574
575			void Toupper(string &s)
576			{
577			unsigned int i;
578			for (i = 0;i < s.size();i++)
579			s[i] = toupper(s[i]);
580			}
581
582			void Tolower(string &s)
583			{
584			unsigned int i;
585			for (i = 0;i < s.size();i++)
586			s[i] = tolower(s[i]);
587			}
588
589			///////////////////////////////////////////////////////////////////////
590			OBResidueData::OBResidueData()
591			{
592			_init = false;
593			_dir = BABEL_DATADIR;
594			_envvar = "BABEL_DATADIR";
595			_filename = "resdata.txt";
596			_subdir = "data";
597			_dataptr = ResidueData;
598			}
599
600			bool OBResidueData::AssignBonds(OBMol &mol,OBBitVec &bv)
601			{
602			OBAtom a1,a2;
603			OBResidue r1,r2;
604			vector<OBNodeBase*>::iterator i,j;
605			vector3 v;
606
607			int bo;
608			unsigned int skipres=0;
609			string rname = "";
610			//assign residue bonds
611			for (a1 = mol.BeginAtom(i);a1;a1 = mol.NextAtom(i))
612			{
613			r1 = a1->GetResidue();
614			if (skipres && r1->GetNum() == skipres)
615			continue;
616
617			if (r1->GetName() != rname)
618			{
619			skipres = SetResName(r1->GetName()) ? 0 : r1->GetNum();
620			rname = r1->GetName();
621			}
622			//assign bonds for each atom
623			for (j=i,a2 = mol.NextAtom(j);a2;a2 = mol.NextAtom(j))
624			{
625			r2 = a2->GetResidue();
626			if (r1->GetNum() != r2->GetNum())
627			break;
628			if (r1->GetName() != r2->GetName())
629			break;
630
631			if ((bo = LookupBO(r1->GetAtomID(a1),r2->GetAtomID(a2))))
632			{
633			v = a1->GetVector() - a2->GetVector();
634			if (v.length_2() < 3.5) //check by distance
635			mol.AddBond(a1->GetIdx(),a2->GetIdx(),bo);
636			}
637			}
638			}
639
640			int hyb;
641			string type;
642
643			//types and hybridization
644			for (a1 = mol.BeginAtom(i);a1;a1 = mol.NextAtom(i))
645			{
646			if (a1->IsOxygen() && !a1->GetValence())
647			{
648			a1->SetType("O3");
649			continue;
650			}
651
652			if (a1->IsHydrogen())
653			{
654			a1->SetType("H");
655			continue;
656			}
657
658			r1 = a1->GetResidue();
659			if (skipres && r1->GetNum() == skipres)
660			continue;
661
662			if (r1->GetName() != rname)
663			{
664			skipres = SetResName(r1->GetName()) ? 0 : r1->GetNum();
665			rname = r1->GetName();
666			}
667
668			//***valence rule for O-
669			if (a1->IsOxygen() && a1->GetValence() == 1)
670			{
671			OBBond *bond;
672			bond = (OBBond)(a1->BeginBonds());
673			if (bond->GetBO() == 2)
674			{
675			a1->SetType("O2");
676			a1->SetHyb(2);
677			}
678			if (bond->GetBO() == 1)
679			{
680			a1->SetType("O-");
681			a1->SetHyb(3);
682			a1->SetFormalCharge(-1);
683			}
684			}
685			else
686			if (LookupType(r1->GetAtomID(a1),type,hyb))
687			{
688			a1->SetType(type);
689			a1->SetHyb(hyb);
690			}
691			else // try to figure it out by bond order ???
692			{}
693			}
694
695			return(true);
696			}
697
698			void OBResidueData::ParseLine(const char *buffer)
699			{
700			int bo;
701			string s;
702			vector<string> vs;
703
704			if (buffer[0] == '#')
705			return;
706
707			tokenize(vs,buffer);
708			if (!vs.empty())
709			{
710			if (vs[0] == "BOND")
711			{
712			s = (vs[1] < vs[2]) ? vs[1] + " " + vs[2] :
713			vs[2] + " " + vs[1];
714			bo = atoi(vs[3].c_str());
715			_vtmp.push_back(pair<string,int> (s,bo));
716			}
717
718			if (vs[0] == "ATOM" && vs.size() == 4)
719			{
720			_vatmtmp.push_back(vs[1]);
721			_vatmtmp.push_back(vs[2]);
722			_vatmtmp.push_back(vs[3]);
723			}
724
725			if (vs[0] == "RES")
726			_resname.push_back(vs[1]);
727
728			if (vs[0]== "END")
729			{
730			_resatoms.push_back(_vatmtmp);
731			_resbonds.push_back(_vtmp);
732			_vtmp.clear();
733			_vatmtmp.clear();
734			}
735			}
736			}
737
738			bool OBResidueData::SetResName(const string &s)
739			{
740			unsigned int i;
741			for (i = 0;i < _resname.size();i++)
742			if (_resname[i] == s)
743			{
744			_resnum = i;
745			return(true);
746			}
747
748			_resnum = -1;
749			return(false);
750			}
751
752			int OBResidueData::LookupBO(const string &s)
753			{
754			if (_resnum == -1)
755			return(0);
756
757			unsigned int i;
758			for (i = 0;i < _resbonds[_resnum].size();i++)
759			if (_resbonds[_resnum][i].first == s)
760			return(_resbonds[_resnum][i].second);
761
762			return(0);
763			}
764
765			int OBResidueData::LookupBO(const string &s1, const string &s2)
766			{
767			if (_resnum == -1)
768			return(0);
769			string s;
770
771			s = (s1 < s2) ? s1 + " " + s2 : s2 + " " + s1;
772
773			unsigned int i;
774			for (i = 0;i < _resbonds[_resnum].size();i++)
775			if (_resbonds[_resnum][i].first == s)
776			return(_resbonds[_resnum][i].second);
777
778			return(0);
779			}
780
781			bool OBResidueData::LookupType(const string &atmid,string &type,int &hyb)
782			{
783			if (_resnum == -1)
784			return(false);
785
786			string s;
787			vector<string>::iterator i;
788
789			for (i = _resatoms[_resnum].begin();i != _resatoms[_resnum].end();i+=3)
790			if (atmid == *i)
791			{
792			i++;
793			type = *i;
794			i++;
795			hyb = atoi((*i).c_str());
796			return(true);
797			}
798
799			return(false);
800			}
801
802			void OBGlobalDataBase::Init()
803			{
804			if (_init)
805			return;
806			_init = true;
807
808			char buffer[BUFF_SIZE],subbuffer[BUFF_SIZE];
809			ifstream ifs1, ifs2, ifs3, ifs4, *ifsP;
810			// First, look for an environment variable
811			if (getenv(_envvar.c_str()) != NULL)
812			{
813			strcpy(buffer,getenv(_envvar.c_str()));
814			strcat(buffer,FILE_SEP_CHAR);
815
816			if (!_subdir.empty())
817			{
818			strcpy(subbuffer,buffer);
819			strcat(subbuffer,_subdir.c_str());
820			strcat(subbuffer,FILE_SEP_CHAR);
821			}
822
823			strcat(buffer,(char*)_filename.c_str());
824			strcat(subbuffer,(char*)_filename.c_str());
825
826			ifs1.open(subbuffer);
827			ifsP= &ifs1;
828			if (!(*ifsP))
829			{
830			ifs2.open(buffer);
831			ifsP = &ifs2;
832			}
833			}
834			// Then, check the configured data directory
835			else // if (!(*ifsP))
836			{
837			strcpy(buffer,_dir.c_str());
838			strcat(buffer,FILE_SEP_CHAR);
839
840			strcpy(subbuffer,buffer);
841			strcat(subbuffer,BABEL_VERSION);
842			strcat(subbuffer,FILE_SEP_CHAR);
843			strcat(subbuffer,(char*)_filename.c_str());
844
845			strcat(buffer,(char*)_filename.c_str());
846
847			ifs3.open(subbuffer);
848			ifsP= &ifs3;
849			if (!(*ifsP))
850			{
851			ifs4.open(buffer);
852			ifsP = &ifs4;
853			}
854			}
855
856			if ((*ifsP))
857			{
858			while(ifsP->getline(buffer,BUFF_SIZE))
859			ParseLine(buffer);
860			}
861
862			else
863			// If all else fails, use the compiled in values
864			if (_dataptr)
865			{
866			const char p1,p2;
867			for (p1 = p2 = _dataptr;*p2 != '\0';p2++)
868			if (*p2 == '\n')
869			{
870			strncpy(buffer, p1, (p2 - p1));
871			buffer[(p2 - p1)] = '\0';
872			ParseLine(buffer);
873			p1 = ++p2;
874			}
875			}
876			else
877			{
878			string s = "Unable to open data file '";
879			s += _filename;
880			s += "'";
881			obErrorLog.ThrowError(__FUNCTION__, s, obWarning);
882			}
883
884			if (ifs1)
885			ifs1.close();
886			if (ifs2)
887			ifs2.close();
888			if (ifs3)
889			ifs3.close();
890			if (ifs4)
891			ifs4.close();
892
893			if (GetSize() == 0)
894			{
895			string s = "Cannot initialize database '";
896			s += _filename;
897			s += "' which may cause further errors.";
898			obErrorLog.ThrowError(__FUNCTION__, "Cannot initialize database", obWarning);
899			}
900
901			}
902
903			} // end namespace OpenBabel
904
905			//! \file data.cpp
906			//! \brief Global data and resource file parsers.