void s_Check(const CBioseq& seq)
{
_ASSERT(!seq.GetId().empty());
const CSeq_inst& inst = seq.GetInst();
const string& seqdata = inst.GetSeq_data().GetIupacna().Get();
_ASSERT(seqdata.size() == inst.GetLength());
ITERATE ( string, i, seqdata ) {
_ASSERT(*i >= 'A' && *i <= 'Z');
}
bool
CBlastBioseqMaker::IsEmptyBioseq(const CBioseq& bioseq)
{
if (bioseq.CanGetInst()) {
const CSeq_inst& inst = bioseq.GetInst();
return (inst.GetRepr() == CSeq_inst::eRepr_raw &&
inst.CanGetMol() &&
inst.CanGetLength() &&
inst.CanGetSeq_data() == false);
}
return false;
}
CSeqVector::CSeqVector(const CBioseq& bioseq,
CScope* scope,
EVectorCoding coding, ENa_strand strand)
: m_Scope(scope),
m_SeqMap(CSeqMap::CreateSeqMapForBioseq(bioseq)),
m_Strand(strand),
m_Coding(CSeq_data::e_not_set)
{
m_Size = m_SeqMap->GetLength(scope);
m_Mol = bioseq.GetInst().GetMol();
SetCoding(coding);
}
bool CSeqAnnotFromFasta::PurgeNonAlphaFromSequence(CBioseq& bioseq) {
bool result = false;
CSeq_inst::TLength newLength;
string originalSequence;
// CRef< CBioseq > bioseq;
// if (cd.GetBioseqForIndex(index, bioseq)) {
if (bioseq.GetInst().IsSetSeq_data()) {
CSeq_data& seqData = bioseq.SetInst().SetSeq_data();
if (seqData.IsNcbieaa()) {
originalSequence = seqData.SetNcbieaa().Set();
} else if (seqData.IsIupacaa()) {
originalSequence = seqData.SetIupacaa().Set();
} else if (seqData.IsNcbistdaa()) {
std::vector < char >& vec = seqData.SetNcbistdaa().Set();
NcbistdaaToNcbieaaString(vec, &originalSequence);
}
if (PurgeNonAlpha(originalSequence)) {
// if (originalSequence.length() > 0 && find_if(originalSequence.begin(), originalSequence.end(), isNotAlpha) != originalSequence.end()) {
// originalSequence.erase(remove_if(originalSequence.begin(), originalSequence.end(), isNotAlpha), originalSequence.end());
// _TRACE("after remove non-alpha: \n" << originalSequence);
seqData.Select(CSeq_data::e_Ncbieaa);
seqData.SetNcbieaa().Set(originalSequence);
result = true;
}
newLength = originalSequence.length();
bioseq.SetInst().SetLength(newLength);
}
// }
return result;
}
int CReadBlastApp::AnalyzeSeqsViaBioseqs(CBioseq& left, CBioseq& right)
{
/*
if(PrintDetails()) NcbiCerr << "AnalyzeSeqsViaBioseqs(left, right): "
<< GetStringDescr(left) << ", " << GetStringDescr(right) << NcbiEndl;
*/
if(is_prot_entry(left) && !is_prot_entry(right))
{
// if(PrintDetails()) NcbiCerr << "AnalyzeSeqsViaBioseqs(left, right): going for overlaps\n";
overlaps_prot_na(left, right.GetAnnot());
}
return -1;
}
Sequence::Sequence(const CBioseq& bioseq) :
status(CAV_ERROR_SEQUENCES), bioseqASN(&bioseq), seqIDs(bioseq.GetId()), mmdbLink(NOT_SET)
{
// try to get description from title or compound
if (bioseq.IsSetDescr()) {
CSeq_descr::Tdata::const_iterator d, de = bioseq.GetDescr().Get().end();
for (d=bioseq.GetDescr().Get().begin(); d!=de; ++d) {
if (d->GetObject().IsTitle()) {
description = d->GetObject().GetTitle();
break;
} else if (d->GetObject().IsPdb() && d->GetObject().GetPdb().GetCompound().size() > 0) {
description = d->GetObject().GetPdb().GetCompound().front();
break;
}
}
}
// get link to MMDB id - mainly for CDD's where Biostrucs have to be loaded separately
if (bioseq.IsSetAnnot()) {
CBioseq::TAnnot::const_iterator a, ae = bioseq.GetAnnot().end();
for (a=bioseq.GetAnnot().begin(); a!=ae; ++a) {
if (a->GetObject().GetData().IsIds()) {
CSeq_annot::C_Data::TIds::const_iterator i, ie = a->GetObject().GetData().GetIds().end();
for (i=a->GetObject().GetData().GetIds().begin(); i!=ie; ++i) {
if (i->GetObject().IsGeneral() &&
i->GetObject().GetGeneral().GetDb() == "mmdb" &&
i->GetObject().GetGeneral().GetTag().IsId()) {
mmdbLink = i->GetObject().GetGeneral().GetTag().GetId();
break;
}
}
if (i != ie) break;
}
}
}
if (mmdbLink != NOT_SET)
ERR_POST_X(3, Info << "sequence " << GetTitle() << " is from MMDB id " << mmdbLink);
// get sequence string
if (bioseq.GetInst().GetRepr() == CSeq_inst::eRepr_raw && bioseq.GetInst().IsSetSeq_data()) {
// protein formats
if (bioseq.GetInst().GetSeq_data().IsNcbieaa()) {
sequenceString = bioseq.GetInst().GetSeq_data().GetNcbieaa().Get();
} else if (bioseq.GetInst().GetSeq_data().IsIupacaa()) {
sequenceString = bioseq.GetInst().GetSeq_data().GetIupacaa().Get();
} else if (bioseq.GetInst().GetSeq_data().IsNcbistdaa()) {
StringFromStdaa(bioseq.GetInst().GetSeq_data().GetNcbistdaa().Get(), &sequenceString);
}
// nucleotide formats
else if (bioseq.GetInst().GetSeq_data().IsIupacna()) {
sequenceString = bioseq.GetInst().GetSeq_data().GetIupacna().Get();
} else if (bioseq.GetInst().GetSeq_data().IsNcbi4na()) {
StringFrom4na(bioseq.GetInst().GetSeq_data().GetNcbi4na().Get(), &sequenceString,
(bioseq.GetInst().GetMol() == CSeq_inst::eMol_dna));
} else if (bioseq.GetInst().GetSeq_data().IsNcbi8na()) { // same repr. for non-X as 4na
StringFrom4na(bioseq.GetInst().GetSeq_data().GetNcbi8na().Get(), &sequenceString,
(bioseq.GetInst().GetMol() == CSeq_inst::eMol_dna));
} else if (bioseq.GetInst().GetSeq_data().IsNcbi2na()) {
StringFrom2na(bioseq.GetInst().GetSeq_data().GetNcbi2na().Get(), &sequenceString,
(bioseq.GetInst().GetMol() == CSeq_inst::eMol_dna));
if (bioseq.GetInst().IsSetLength() && bioseq.GetInst().GetLength() < sequenceString.length())
sequenceString.resize(bioseq.GetInst().GetLength());
}
else {
ERR_POST_X(4, Critical << "Sequence::Sequence() - sequence " << GetTitle()
<< ": confused by sequence string format");
return;
}
if (bioseq.GetInst().IsSetLength() && bioseq.GetInst().GetLength() != sequenceString.length()) {
ERR_POST_X(5, Critical << "Sequence::Sequence() - sequence string length mismatch");
return;
}
} else {
ERR_POST_X(6, Critical << "Sequence::Sequence() - sequence " << GetTitle()
<< ": confused by sequence representation");
return;
}
status = CAV_SUCCESS;
}
//
// CreateConsensus()
//
// compute a consensus sequence given a particular alignment
// the rules for a consensus are:
// - a segment is consensus gap if > 50% of the sequences are gap at this
// segment. 50% exactly is counted as sequence
// - for a segment counted as sequence, for each position, the most
// frequently occurring base is counted as consensus. in the case of
// a tie, the consensus is considered muddied, and the consensus is
// so marked
//
CRef<CDense_seg>
CAlnVec::CreateConsensus(int& consensus_row, CBioseq& consensus_seq,
const CSeq_id& consensus_id) const
{
consensus_seq.Reset();
if ( !m_DS || m_NumRows < 1) {
return CRef<CDense_seg>();
}
bool isNucleotide = GetBioseqHandle(0).IsNucleotide();
size_t i;
size_t j;
// temporary storage for our consensus
vector<string> consens(m_NumSegs);
CreateConsensus(consens);
//
// now, create a new CDense_seg
// we create a new CBioseq for our data and
// copy the contents of the CDense_seg
//
string data;
TSignedSeqPos total_bases = 0;
CRef<CDense_seg> new_ds(new CDense_seg());
new_ds->SetDim(m_NumRows + 1);
new_ds->SetNumseg(m_NumSegs);
new_ds->SetLens() = m_Lens;
new_ds->SetStarts().reserve(m_Starts.size() + m_NumSegs);
if ( !m_Strands.empty() ) {
new_ds->SetStrands().reserve(m_Strands.size() +
m_NumSegs);
}
for (i = 0; i < consens.size(); ++i) {
// copy the old entries
for (j = 0; j < (size_t)m_NumRows; ++j) {
int idx = i * m_NumRows + j;
new_ds->SetStarts().push_back(m_Starts[idx]);
if ( !m_Strands.empty() ) {
new_ds->SetStrands().push_back(m_Strands[idx]);
}
}
// add our new entry
// this places the consensus as the last sequence
// it should preferably be the first, but this would mean adjusting
// the bioseq handle and seqvector caches, and all row numbers would
// shift
if (consens[i].length() != 0) {
new_ds->SetStarts().push_back(total_bases);
} else {
new_ds->SetStarts().push_back(-1);
}
if ( !m_Strands.empty() ) {
new_ds->SetStrands().push_back(eNa_strand_unknown);
}
total_bases += consens[i].length();
data += consens[i];
}
// copy our IDs
for (i = 0; i < m_Ids.size(); ++i) {
new_ds->SetIds().push_back(m_Ids[i]);
}
// now, we construct a new Bioseq
{{
// sequence ID
CRef<CSeq_id> id(new CSeq_id());
id->Assign(consensus_id);
consensus_seq.SetId().push_back(id);
new_ds->SetIds().push_back(id);
// add a description for this sequence
CSeq_descr& desc = consensus_seq.SetDescr();
CRef<CSeqdesc> d(new CSeqdesc);
desc.Set().push_back(d);
d->SetComment("This is a generated consensus sequence");
// the main one: Seq-inst
CSeq_inst& inst = consensus_seq.SetInst();
inst.SetRepr(CSeq_inst::eRepr_raw);
inst.SetMol(isNucleotide ? CSeq_inst::eMol_na : CSeq_inst::eMol_aa);
inst.SetLength(data.length());
CSeq_data& seq_data = inst.SetSeq_data();
if (isNucleotide) {
CIUPACna& na = seq_data.SetIupacna();
na = CIUPACna(data);
} else {
CIUPACaa& aa = seq_data.SetIupacaa();
aa = CIUPACaa(data);
}
}}
consensus_row = new_ds->GetIds().size() - 1;
return new_ds;
}
