/// Performs sanity checks to make sure that the sequence requested is of
/// the expected type. If the tests fail, an exception is thrown.
/// @param id Sequence id for this sequence [in]
void x_ValidateMoleculeType(CConstRef<CSeq_id> id)
{
_ASSERT(m_BioseqMaker.NotEmpty());
if (id.Empty())
{
NCBI_THROW(CInputException, eInvalidInput,
"Empty SeqID passed to the molecule type validation");
}
bool isProtein = m_BioseqMaker->IsProtein(id);
if (!isProtein && m_ReadProteins)
{
NCBI_THROW(CInputException, eSequenceMismatch,
"GI/accession/sequence mismatch: protein input required but nucleotide provided");
}
if (isProtein && !m_ReadProteins)
{
NCBI_THROW(CInputException, eSequenceMismatch,
"GI/accession/sequence mismatch: nucleotide input required but protein provided");
}
if (!isProtein) // Never seen a virtual protein sequence.
{
if (m_BioseqMaker->HasSequence(id) == false)
{
string message = "No sequence available for " + id->AsFastaString();
NCBI_THROW(CInputException, eInvalidInput, message);
}
}
}
pair<double, bool> CScoreUniqSeqCoverage::MakeScore(CBioseq_Handle const& query_handle, vector<CSeq_align const*>::const_iterator begin, vector<CSeq_align const*>::const_iterator end)
{
CConstRef<CBioseq> bioseq = query_handle.GetCompleteBioseq();
unsigned int qlen = 0;
if ( !bioseq.Empty() && bioseq->IsSetLength()) {
qlen = bioseq->GetLength();
}
if ( !qlen ) {
return make_pair(0, false);
}
bool isDenDiag = ( (*begin)->GetSegs().Which() == CSeq_align::C_Segs::e_Dendiag) ?
true : false;
CRangeCollection<TSeqPos> subj_rng_coll((*begin)->GetSeqRange(1));
CRange<TSeqPos> q_rng((*begin)->GetSeqRange(0));
CRangeCollection<TSeqPos> query_rng_coll(s_FixMinusStrandRange(q_rng));
for( ++begin; begin != end; ++begin ) {
const CRange<TSeqPos> align_subj_rng((*begin)->GetSeqRange(1));
// subject range should always be on the positive strand
assert(align_subj_rng.GetTo() > align_subj_rng.GetFrom());
CRangeCollection<TSeqPos> coll(align_subj_rng);
coll.Subtract(subj_rng_coll);
if ( coll.empty() ) {
continue;
}
if(coll[0] == align_subj_rng) {
CRange<TSeqPos> query_rng ((*begin)->GetSeqRange(0));
query_rng_coll += s_FixMinusStrandRange(query_rng);
subj_rng_coll += align_subj_rng;
}
else {
ITERATE (CRangeCollection<TSeqPos>, uItr, coll) {
CRange<TSeqPos> query_rng;
const CRange<TSeqPos> & subj_rng = (*uItr);
CRef<CSeq_align> densegAln;
if ( isDenDiag) {
densegAln = CreateDensegFromDendiag(**begin);
}
CAlnMap map( (isDenDiag) ? densegAln->GetSegs().GetDenseg() : (*begin)->GetSegs().GetDenseg());
TSignedSeqPos subj_aln_start = map.GetAlnPosFromSeqPos(1,subj_rng.GetFrom());
TSignedSeqPos subj_aln_end = map.GetAlnPosFromSeqPos(1,subj_rng.GetTo());
query_rng.SetFrom(map.GetSeqPosFromAlnPos(0,subj_aln_start));
query_rng.SetTo(map.GetSeqPosFromAlnPos(0,subj_aln_end));
query_rng_coll += s_FixMinusStrandRange(query_rng);
subj_rng_coll += subj_rng;
}
}
}
bool CLocation_constraint :: Match(const CSeq_feat& feat, CConstRef <CSeq_feat> feat_to, CConstRef <CBioseq> feat_bioseq) const
{
if (x_IsLocationConstraintEmpty()) {
return true;
}
const CSeq_loc& feat_loc = feat.GetLocation();
if (GetStrand() != eStrand_constraint_any) {
if (feat_bioseq.Empty()) {
return false;
}
else if (feat_bioseq->IsAa()) {
if (feat_to.Empty()) { // when feat is product, feat_to points to cds
return false;
}
else if (!x_DoesStrandMatchConstraint (feat_to->GetLocation())) {
return false;
}
}
else if (!x_DoesStrandMatchConstraint (feat_loc)) {
return false;
}
}
if (!x_DoesBioseqMatchSequenceType(feat_bioseq, GetSeq_type())) {
return false;
}
if (!x_DoesLocationMatchPartialnessConstraint (feat_loc)) {
return false;
}
if (!x_DoesLocationMatchTypeConstraint (feat_loc)) {
return false;
}
if (!x_DoesLocationMatchDistanceConstraint(feat_bioseq, feat_loc)) {
return false;
}
return true;
};
bool CLocation_constraint :: x_DoesLocationMatchDistanceConstraint(CConstRef <CBioseq> bioseq, const CSeq_loc& loc) const
{
if (!CanGetEnd5() && !CanGetEnd3()) {
return true;
}
unsigned pos = loc.GetStop(eExtreme_Positional);
int pos2;
if (bioseq.NotEmpty()) {
pos2 = (bioseq->IsSetLength() ? bioseq->GetLength() : 0) - pos - 1;
}
if (loc.GetStrand() == eNa_strand_minus) {
if (CanGetEnd5()) {
if (bioseq.Empty()) {
return false;
}
else {
if (!GetEnd5().Match(pos2)) {
return false;
}
}
}
if (CanGetEnd3()) {
return GetEnd3().Match(pos);
}
}
else
{
if (CanGetEnd5() && !GetEnd5().Match(pos)) {
return false;
}
if (CanGetEnd3()) {
if (bioseq.Empty()) {
return false;
}
return GetEnd3().Match(pos2);
}
}
return true;
};
void
CCmdLineBlastXML2ReportData::x_InitCommon(
const CSearchResults & results,
CConstRef<CBlastOptions> opts)
{
if(opts.Empty()) {
NCBI_THROW(CException, eUnknown, "blastxml2: Empty blast options");
}
if(m_Scope.Empty()) {
NCBI_THROW(CException, eUnknown, "blastxml2: Empty scope");
}
x_FillScoreMatrix(m_Options->GetMatrixName());
string resolved = SeqDB_ResolveDbPath("taxdb.bti");
if(!resolved.empty()) {
m_TaxDBFound = true;
}
m_isIterative = opts->IsIterativeSearch();
}
pair<double, bool> CScoreSeqCoverage::MakeScore(CBioseq_Handle const& query_handle, vector<CSeq_align const*>::const_iterator begin, vector<CSeq_align const*>::const_iterator end)
{
CConstRef<CBioseq> bioseq = query_handle.GetCompleteBioseq();
unsigned int qlen = 0;
if ( !bioseq.Empty() && bioseq->IsSetLength()) {
qlen = bioseq->GetLength();
}
if ( !qlen ) {
return make_pair(0, false);
}
// Subject coverage score
CRangeCollection<TSeqPos> range_coll;
for ( ; begin != end; ++begin ) {
CRange<TSeqPos> range = (*begin)->GetSeqRange(0);
s_FixMinusStrandRange(range);
range_coll += range;
}
double score = ( 100.0 * range_coll.GetCoveredLength() ) / qlen;
return make_pair(score, true);
}
