static void fillFrag(
::libmaus::bambam::BamAlignment const & p,
::libmaus::bambam::BamHeader const & header,
::libmaus::bambam::ReadEnds & RE
)
{
RE.libraryId = p.getLibraryId(header);
RE.read1Sequence = p.getRefIDChecked() + 1;
RE.read1Coordinate = p.getCoordinate() + 1;
RE.orientation = p.isReverse() ? ::libmaus::bambam::ReadEnds::R : ::libmaus::bambam::ReadEnds::F;
RE.read1IndexInFile = p.getRank();
RE.score = p.getScore();
if ( p.isPaired() && (!p.isMateUnmap()) )
RE.read2Sequence = p.getNextRefIDChecked() + 1;
char const * readname = p.getName();
char const * readnamee = readname + (p.getLReadName()-1);
int cnt[2] = { 0,0 };
for ( char const * c = readname; c != readnamee; ++c )
cnt [ (static_cast<int>(*c) - ':') == 0 ] ++;
bool const rnparseok = (cnt[1] == 4);
// parse tile, x, y
if ( rnparseok )
{
uint8_t const * sem[4];
uint8_t const ** psem = &sem[0];
for ( uint8_t const * c = reinterpret_cast<uint8_t const *>(readname); c != reinterpret_cast<uint8_t const *>(readnamee); ++c )
if ( *c == ':' )
*(psem++) = c+1;
uint8_t const * t = sem[1];
while ( D[*t] )
{
RE.tile *= 10;
RE.tile += *(t++)-'0';
}
RE.tile += 1;
t = sem[2];
while ( D[*t] )
{
RE.x *= 10;
RE.x += *(t++)-'0';
}
t = sem[3];
while ( D[*t] )
{
RE.y *= 10;
RE.y += *(t++)-'0';
}
}
int64_t const rg = p.getReadGroupId(header);
RE.readGroup = rg + 1;
}
bool strip12(libmaus::bambam::BamAlignment & algn)
{
char const * name = algn.getName();
char const * u1 = name;
while ( *u1 && *u1 != '_' )
++u1;
if ( ! *u1 )
return true;
else
{
bool ok = true;
uint64_t ranka = 0;
for ( char const * t1 = name; t1 != u1; ++t1 )
{
ranka *= 10;
ranka += ((*t1)-'0');
ok = ok && isdigit(*t1);
}
int const read1 = algn.isRead1() ? 1 : 0;
int const read2 = algn.isRead2() ? 1 : 0;
if ( (read1+read2 != 1) || (!ok) )
{
return true;
}
else
{
std::ostringstream upnamestr;
upnamestr << (u1+1);
std::string const upname = upnamestr.str();
algn.replaceName(upname.begin(),upname.size());
return true;
}
}
}
bool checkCigarValid(
::libmaus::bambam::BamAlignment const & alignment,
::libmaus::bambam::BamHeader const & bamheader,
::libmaus::autoarray::AutoArray < ::libmaus::autoarray::AutoArray<uint8_t>::unique_ptr_type > const & text
)
{
if ( alignment.isUnmap() )
return true;
if ( ! alignment.isCigarLengthConsistent() )
{
std::cerr << "[E] inconsistent cigar " << alignment.getCigarString() << " for " << alignment.getName() << std::endl;
return false;
}
if ( alignment.getRefID() < 0 || alignment.getRefID() >= static_cast<int64_t>(bamheader.chromosomes.size()) )
{
std::cerr << "[E] reference id " << alignment.getRefID() << " out of range for " << alignment.getName() << std::endl;
return false;
}
::libmaus::autoarray::AutoArray<uint8_t> const & ctext = *(text[alignment.getRefID()]);
int64_t refpos = alignment.getPos();
int64_t seqpos = 0;
bool alok = true;
std::string const read = alignment.getRead();
for ( uint64_t i = 0; alok && i < alignment.getNCigar(); ++i )
{
char const cop = alignment.getCigarFieldOpAsChar(i);
int64_t const clen = alignment.getCigarFieldLength(i);
switch ( cop )
{
// match/mismatch, increment both
case '=':
case 'X':
case 'M':
{
for ( int64_t j = 0; alok && j < clen; ++j, ++refpos, ++ seqpos )
{
if ( refpos < 0 || refpos >= static_cast<int64_t>(ctext.size()) )
{
std::cerr << "[E] " << cop << " operation outside of chromosome coordinate range " << " for " << alignment.getName() << std::endl;
alok = false;
}
else if ( seqpos >= alignment.getLseq() )
{
std::cerr << "[E] " << cop << " operation outside of sequence coordinate range " << " for " << alignment.getName() << std::endl;
alok = false;
}
else if ( cop == '=' && toupper(ctext[refpos]) != toupper(read[seqpos]) )
{
std::cerr << "[E] " << cop << " operation but mismatch between reference and query." << std::endl;
alok = false;
}
else if ( cop == 'X' && toupper(ctext[refpos]) == toupper(read[seqpos]) )
{
std::cerr << "[E] " << cop << " operation but mismatch between reference and query." << std::endl;
alok = false;
}
}
break;
}
// insert into reference, increment seq
case 'P':
case 'I':
{
for ( int64_t j = 0; alok && j < clen; ++j, ++seqpos )
{
if ( seqpos >= alignment.getLseq() )
{
std::cerr << "[E] " << cop << " operation outside of sequence coordinate range " << " for " << alignment.getName() << std::endl;
alok = false;
}
}
break;
}
// delete from reference, increment ref
case 'D':
{
for ( int64_t j = 0; alok && j < clen; ++j, ++refpos )
{
if ( refpos < 0 || refpos >= static_cast<int64_t>(ctext.size()) )
{
std::cerr << "[E] " << cop << " operation outside of reference coordinate range " << " for " << alignment.getName() << std::endl;
alok = false;
}
}
break;
}
// soft clipping, increment seq
case 'S':
{
for ( int64_t j = 0; alok && j < clen; ++j, ++seqpos )
{
if ( seqpos >= alignment.getLseq() )
{
std::cerr << "[E] " << cop << " operation outside of sequence coordinate range " << " for " << alignment.getName() << std::endl;
alok = false;
}
}
break;
}
// hard clipping, do nothing
case 'H':
{
break;
}
// skip region in reference, increment ref
case 'N':
{
for ( int64_t j = 0; alok && j < clen; ++j, ++refpos )
{
if ( refpos < 0 || refpos >= static_cast<int64_t>(ctext.size()) )
{
std::cerr << "[E] " << cop << " operation outside of reference coordinate range " << " for " << alignment.getName() << std::endl;
alok = false;
}
}
break;
}
}
}
return alok;
}