/** Function get_seeds()
*
* Produce seeds: rID-> (s0, s1, s2...) for each fragment given two
* fq files f and f2
*/
void get_seeds (ii64vec_t& list_seeds, const std::string& f,
const std::string& f2, int seed_len, int batch, bool silent) {
std::ifstream fh, fh2;
xny::openfile<std::ifstream>(fh, f);
xny::openfile<std::ifstream>(fh2, f2);
bio::fastq_input_iterator<> fq(fh), end, fq2(fh2);
int total_read_pairs = 0;
strvec_t pairs;
while (fq != end && fq2 != end) {
add_fq_reads_only (pairs, batch/2, fq, end);
add_fq_reads_only (pairs, batch/2, fq2, end);
generate_seeds (list_seeds, pairs, seed_len);
total_read_pairs += pairs.size()/2;
pairs.clear();
} // while
if (!silent) {
std::cout << "\t\ttotal frags: " << total_read_pairs << "\n";
}
xny::closefile(fh);
xny::closefile(fh2);
} // get_seeds
/** Function get_super_sketches
*
*/
void get_super_sketches (std::vector<sketch_t>& super_sketches,
const std::string& f, const std::string& f2, xny::sketch_list& slistgen,
xny::super_sketch& ssgen, jaz::murmur264& hashfunc, int batch, bool silent) {
std::ifstream fh, fh2;
xny::openfile<std::ifstream>(fh, f);
xny::openfile<std::ifstream>(fh2, f2);
bio::fastq_input_iterator<> fq(fh), end, fq2(fh2);
int total_read_pairs = 0;
strvec_t pairs;
while (fq != end && fq2 != end) {
add_fq_reads_only (pairs, batch/2, fq, end);
add_fq_reads_only (pairs, batch/2, fq2, end);
generate_super_sketches (super_sketches, pairs, slistgen,
ssgen, hashfunc);
total_read_pairs += pairs.size()/2;
pairs.clear();
} // while
if (!silent) {
std::cout << "\t\t\ttotal pairs, super_sketches: " << total_read_pairs
<< ", " << super_sketches.size() << "\n";
}
xny::closefile(fh);
xny::closefile(fh2);
} // get_super_sketches
/** Function clean_dupl_frag ()
*
*/
void clean_dupl_frag (const std::string& ifq, const std::string& ifq2,
std::ofstream& ofhfq, std::ofstream& ofhfq2, const iset_t& duplIDs,
xny::low_complexity& lc, int batch){
std::ifstream ifhfq, ifhfq2;
xny::openfile<std::ifstream>(ifhfq, ifq);
xny::openfile<std::ifstream>(ifhfq2, ifq2);
bio::fastq_input_iterator<> fq(ifhfq), end, fq2(ifhfq2);
int total_read_pairs = 0;
std::vector<fqtuple_t> pairs;
int num_lc = 0;
int fragID = 0;
while (fq != end && fq2 != end) {
add_fq_reads (pairs, batch/2, fq, end);
add_fq_reads (pairs, batch/2, fq2, end);
iset_t low_complex_frag;
check_low_complexity (low_complex_frag, fragID, pairs, lc);
num_lc += low_complex_frag.size();
int fragnum = pairs.size()/2;
for (int i = 0; i < fragnum; ++ i) {
if ( (!duplIDs.count(fragID)) &&
(!low_complex_frag.count(fragID))) { // output
ofhfq << "@" << std::get<0>(pairs[i]) << "\n";
ofhfq << std::get<1>(pairs[i]) << "\n";
ofhfq << "+\n";
ofhfq << std::get<2>(pairs[i]) << "\n";
ofhfq2 << "@" << std::get<0>(pairs[i + fragnum]) << "\n";
ofhfq2 << std::get<1>(pairs[i + fragnum]) << "\n";
ofhfq2 << "+\n";
ofhfq2 << std::get<2>(pairs[i + fragnum]) << "\n";
}
++ fragID;
}
total_read_pairs += pairs.size()/2;
pairs.clear();
} // while
std::cout << "\t\tlow complexity fragments: " << num_lc << "\n\n";
xny::closefile(ifhfq);
xny::closefile(ifhfq2);
} // clean_dupl_frag
void BamToFastq::PairedFastq() {
// open the 1st fastq file for writing
ofstream fq1(_fastq1.c_str(), ios::out);
if ( !fq1 ) {
cerr << "Error: The first fastq file (" << _fastq1 << ") could not be opened. Exiting!" << endl;
exit (1);
}
// open the 2nd fastq file for writing
ofstream fq2(_fastq2.c_str(), ios::out);
if ( !fq2 ) {
cerr << "Error: The second fastq file (" << _fastq2 << ") could not be opened. Exiting!" << endl;
exit (1);
}
// open the BAM file
BamReader reader;
reader.Open(_bamFile);
// rip through the BAM file and convert each mapped entry to BEDPE
BamAlignment bam1, bam2;
bool shouldConsumeReads = true;
while (true) {
if (shouldConsumeReads) {
if (!reader.GetNextAlignment(bam1) || !reader.GetNextAlignment(bam2)) break;
} else {
shouldConsumeReads = true;
}
if (bam1.Name != bam2.Name) {
while (bam1.Name != bam2.Name)
{
if (bam1.IsPaired())
{
cerr << "*****WARNING: Query " << bam1.Name
<< " is marked as paired, but its mate does not occur"
<< " next to it in your BAM file. Skipping. " << endl;
}
bam1 = bam2;
if (!reader.GetNextAlignment(bam2)) break;
shouldConsumeReads = false;
}
}
else if (bam1.IsPaired() && bam2.IsPaired()) {
// extract the sequence and qualities for the BAM "query"
string seq1 = bam1.QueryBases;
string qual1 = bam1.Qualities;
string seq2 = bam2.QueryBases;
string qual2 = bam2.Qualities;
if (bam1.IsReverseStrand() == true) {
reverseComplement(seq1);
reverseSequence(qual1);
}
if (bam2.IsReverseStrand() == true) {
reverseComplement(seq2);
reverseSequence(qual2);
}
fq1 << "@" << bam1.Name << "/1" << endl;
fq1 << seq1 << endl;
fq1 << "+" << endl;
fq1 << qual1 << endl;
fq2 << "@" << bam2.Name << "/2" << endl;
fq2 << seq2 << endl;
fq2 << "+" << endl;
fq2 << qual2 << endl;
}
}
reader.Close();
}
void BamToFastq::PairedFastqUseTags() {
// open the 1st fastq file for writing
ofstream fq1(_fastq1.c_str(), ios::out);
if ( !fq1 ) {
cerr << "Error: The first fastq file (" << _fastq1 << ") could not be opened. Exiting!" << endl;
exit (1);
}
// open the 2nd fastq file for writing
ofstream fq2(_fastq2.c_str(), ios::out);
if ( !fq2 ) {
cerr << "Error: The second fastq file (" << _fastq2 << ") could not be opened. Exiting!" << endl;
exit (1);
}
// open the BAM file
BamReader reader;
reader.Open(_bamFile);
// rip through the BAM file and convert each mapped entry to BEDPE
BamAlignment bam1, bam2;
while (reader.GetNextAlignment(bam1)) {
reader.GetNextAlignment(bam2);
if (bam1.Name != bam2.Name) {
while (bam1.Name != bam2.Name)
{
if (bam1.IsPaired())
{
cerr << "*****WARNING: Query " << bam1.Name
<< " is marked as paired, but it's mate does not occur"
<< " next to it in your BAM file. Skipping. " << endl;
}
bam1 = bam2;
reader.GetNextAlignment(bam2);
}
}
else if (bam1.IsPaired() && bam2.IsPaired()) {
// assume the R2 and Q2 tags are on the + strand.
string mateSequence, mateQualities;
bam1.GetTag("R2", mateSequence);
bam1.GetTag("Q2", mateQualities);
string seq1 = bam1.QueryBases;
string qual1 = bam1.Qualities;
if (bam1.IsReverseStrand() == true) {
reverseComplement(seq1);
reverseSequence(qual1);
}
// since the info for both ends are contained in each BAM record,
// we only need to process one of the two records (bam1) in order
// to produce FASTQ entries for both ends.
// NOTE: Assumes that R2 and Q2 have already been rev
// and revcomped if necessary
if (bam1.IsFirstMate() == true) {
// end1
fq1 << "@" << bam1.Name << "/1" << endl;
fq1 << seq1 << endl;
fq1 << "+" << endl;
fq1 << qual1 << endl;
// end2
fq2 << "@" << bam1.Name << "/2" <<endl;
fq2 << mateSequence << endl;
fq2 << "+" << endl;
fq2 << mateQualities << endl;
}
else {
// end 2
fq2 << "@" << bam1.Name << "/2" <<endl;
fq2 << seq1 << endl;
fq2 << "+" << endl;
fq2 << qual1 << endl;
// end 1
fq1 << "@" << bam1.Name << "/1" <<endl;
fq1 << mateSequence << endl;
fq1 << "+" << endl;
fq1 << mateQualities << endl;
}
}
}
reader.Close();
}