static bool FindHELMAtom(std::vector<AtomPDBResidueInfo*> *seq,
AtomPDBResidueInfo *info,
std::string &id, std::string &pos)
{
char buffer[32];
char ch;
const char *ptr = info->getName().c_str();
if (ptr[0]==' ' && ptr[1]=='S' && ptr[2]=='G' && ptr[3]==' ')
ch = '3';
else if (ptr[0]==' ' && ptr[1]=='N' && ptr[2]==' ' && ptr[3]==' ')
ch = '1';
else if (ptr[0]==' ' && ptr[1]=='C' && ptr[2]==' ' && ptr[3]==' ')
ch = '2';
else return false;
int resno = info->getResidueNumber();
for (unsigned int i=1; i<10; i++) {
unsigned int len = (unsigned int)seq[i].size();
for (unsigned int j=0; j<len; j++) {
AtomPDBResidueInfo *targ = seq[i][j];
if (targ->getResidueNumber() == resno &&
targ->getResidueName() == info->getResidueName() &&
targ->getChainId() == info->getChainId() &&
targ->getInsertionCode() == info->getInsertionCode()) {
id = "PEPTIDE";
id += (char)(i+'0');
sprintf(buffer,"%u:R%c",j+1,ch);
pos = buffer;
return true;
}
}
}
return false;
}
static bool StandardPDBDoubleBond(RWMol *mol, Atom *beg, Atom *end)
{
AtomPDBResidueInfo *bInfo = (AtomPDBResidueInfo*)beg->getMonomerInfo();
if (!bInfo || bInfo->getMonomerType() != AtomMonomerInfo::PDBRESIDUE)
return false;
AtomPDBResidueInfo *eInfo = (AtomPDBResidueInfo*)end->getMonomerInfo();
if (!eInfo || eInfo->getMonomerType() != AtomMonomerInfo::PDBRESIDUE)
return false;
if (!SamePDBResidue(bInfo,eInfo))
return false;
if (bInfo->getIsHeteroAtom() || eInfo->getIsHeteroAtom())
return false;
const char *ptr = bInfo->getResidueName().c_str();
unsigned int rescode = BCNAM(ptr[0],ptr[1],ptr[2]);
ptr = bInfo->getName().c_str();
unsigned int atm1 = BCATM(ptr[0],ptr[1],ptr[2],ptr[3]);
ptr = eInfo->getName().c_str();
unsigned int atm2 = BCATM(ptr[0],ptr[1],ptr[2],ptr[3]);
if (!StandardPDBDoubleBond(rescode,atm1,atm2))
return false;
// Check that neither end already has a double bond
ROMol::OBOND_ITER_PAIR bp;
for (bp=mol->getAtomBonds(beg); bp.first!=bp.second; ++bp.first)
if ((*mol)[*bp.first].get()->getBondType() == Bond::DOUBLE)
return false;
for (bp=mol->getAtomBonds(end); bp.first!=bp.second; ++bp.first)
if ((*mol)[*bp.first].get()->getBondType() == Bond::DOUBLE)
return false;
return true;
}
std::string MolToSequence(const ROMol &mol)
{
std::string result;
for (ROMol::ConstAtomIterator atomIt=mol.beginAtoms();
atomIt!=mol.endAtoms();++atomIt){
const Atom *atom = *atomIt;
AtomPDBResidueInfo *info = (AtomPDBResidueInfo*)(atom->getMonomerInfo());
if (info && info->getMonomerType()==AtomMonomerInfo::PDBRESIDUE &&
info->getName() == " CA ") {
result += getOneLetterCode(info);
}
}
return result;
}
std::string MolToHELM(const ROMol &mol)
{
std::vector<AtomPDBResidueInfo*> seq[10];
std::string result;
bool first = true;
std::string chain;
int id = 1;
/* First pass: Monomers */
for (ROMol::ConstAtomIterator atomIt=mol.beginAtoms();
atomIt!=mol.endAtoms();++atomIt){
const Atom *atom = *atomIt;
AtomPDBResidueInfo *info = (AtomPDBResidueInfo*)(atom->getMonomerInfo());
// We can only write HELM if all atoms have PDB residue information
if (!info || info->getMonomerType()!=AtomMonomerInfo::PDBRESIDUE)
return "";
if (info->getName() == " CA ") {
const char *mono = getHELMMonomer(info);
if (!mono)
return "";
if (first) {
chain = info->getChainId();
result = "PEPTIDE1{";
first = false;
} else if (info->getChainId() != chain) {
// Nine chains should be enough?
if (id == 9)
return "";
id++;
chain = info->getChainId();
result += "}|PEPTIDE";
result += (char)(id+'0');
result += "{";
} else result += ".";
result += mono;
seq[id].push_back(info);
} else if (info->getResidueName() == "NH2" &&
info->getName() == " N ") {
if (first)
return "";
result += ".[am]";
} else if (info->getResidueName() == "ACE" &&
info->getName() == " C ") {
if (first) {
chain = info->getChainId();
result = "PEPTIDE1{[ac]";
first = false;
} else if (info->getChainId() != chain) {
// Nine chains should be enough?
if (id == 9)
return "";
id++;
chain = info->getChainId();
result += "}|PEPTIDE";
result += (char)(id+'0');
result += "{[ac]";
} else return "";
seq[id].push_back(info);
}
}
if (first)
return "";
result += "}$";
first = true;
for (ROMol::ConstBondIterator bondIt=mol.beginBonds();
bondIt!=mol.endBonds(); ++bondIt){
const Bond *bond = *bondIt;
Atom *beg = bond->getBeginAtom();
Atom *end = bond->getEndAtom();
if (!beg || !end)
continue;
AtomPDBResidueInfo *binfo = (AtomPDBResidueInfo*)(beg->getMonomerInfo());
AtomPDBResidueInfo *einfo = (AtomPDBResidueInfo*)(end->getMonomerInfo());
if (!binfo || !einfo)
continue;
// Test if this is an uninteresting intra-residue bond
if (binfo->getResidueNumber() == einfo->getResidueNumber() &&
binfo->getResidueName() == einfo->getResidueName() &&
binfo->getChainId() == einfo->getChainId() &&
binfo->getInsertionCode() == einfo->getInsertionCode())
continue;
if (bond->getBondType() != Bond::SINGLE)
return "";
if (IsEupeptideBond(binfo,einfo))
continue;
if (!IsSupportedHELMBond(binfo,einfo))
return "";
std::string tmp = NameHELMBond(seq,binfo,einfo);
if (tmp.empty())
return "";
if (!first)
result += "|";
else first = false;
result += tmp;
}
result += "$$$";
return result;
}
