int main(int argc, char const ** argv)
{
double startTime = 0;
// -----------------------------------------------------------------------
// Parse command line.
// -----------------------------------------------------------------------
FxSamCoverageOptions options;
seqan::ArgumentParser::ParseResult res = parseArgs(options, argc, argv);
if (res != seqan::ArgumentParser::PARSE_OK)
return res == seqan::ArgumentParser::PARSE_ERROR; // 1 on errors, 0 otherwise
// -----------------------------------------------------------------------
// Show options.
// -----------------------------------------------------------------------
if (options.verbosity >= 1)
{
std::cerr << "____OPTIONS___________________________________________________________________\n"
<< "\n"
<< "VERBOSITY " << options.verbosity << "\n"
<< "GENOME " << options.inGenomePath << "\n"
<< "SAM " << options.inSamPath << "\n"
<< "OUT " << options.outPath << "\n"
<< "WINDOW SIZE " << options.windowSize << "\n";
}
// -----------------------------------------------------------------------
// Load Genome FAI Index
// -----------------------------------------------------------------------
std::cerr << "\n"
<< "___PREPRATION_____________________________________________________________________\n"
<< "\n"
<< "Indexing GENOME file " << options.inGenomePath << " ...";
seqan::FaiIndex faiIndex;
if (build(faiIndex, toCString(options.inGenomePath)) != 0)
{
std::cerr << "Could not build FAI index.\n";
return 1;
}
std::cerr << " OK\n";
// Prepare bins.
seqan::String<seqan::String<BinData> > bins;
resize(bins, numSeqs(faiIndex));
// -----------------------------------------------------------------------
// Compute C+G content
// -----------------------------------------------------------------------
std::cerr << "\n"
<< "___C+G CONTENT COMPUTATION________________________________________________________\n"
<< "\n";
for (unsigned i = 0; i < numSeqs(faiIndex); ++i)
{
std::cerr << "[" << sequenceName(faiIndex, i) << "] ...";
unsigned numBins = (sequenceLength(faiIndex, i) + options.windowSize - 1) / options.windowSize;
resize(bins[i], numBins);
seqan::Dna5String contigSeq;
if (readSequence(contigSeq, faiIndex, i) != 0)
{
std::cerr << "\nERROR: Could not read sequence " << sequenceName(faiIndex, i) << " from file!\n";
return 1;
}
for (unsigned bin = 0; bin < numBins; ++bin)
{
unsigned cgCounter = 0;
unsigned binSize = 0;
bins[i][bin].length = options.windowSize;
if ((bin + 1) * options.windowSize > length(contigSeq))
bins[i][bin].length = length(contigSeq) - bin * options.windowSize;
for (unsigned pos = bin * options.windowSize; pos < length(contigSeq) && pos < (bin + 1) * options.windowSize; ++pos, ++binSize)
cgCounter += (contigSeq[pos] == 'C' || contigSeq[pos] == 'G');
bins[i][bin].cgContent = 1.0 * cgCounter / binSize;
}
std::cerr << "DONE\n";
}
// -----------------------------------------------------------------------
// Compute Coverage
// -----------------------------------------------------------------------
std::cerr << "\n"
<< "___COVERAGE COMPUATATION________________________________________________________\n"
<< "\n"
<< "Computing Coverage...";
seqan::BamStream bamStream(toCString(options.inSamPath));
if (!isGood(bamStream))
{
std::cerr << "Could not open " << options.inSamPath << "!\n";
return 1;
}
seqan::BamAlignmentRecord record;
while (!atEnd(bamStream))
{
if (readRecord(record, bamStream) != 0)
{
std::cerr << "ERROR: Could not read record from BAM file!\n";
return 1;
}
if (hasFlagUnmapped(record) || hasFlagSecondary(record) || record.rId == seqan::BamAlignmentRecord::INVALID_REFID)
continue; // Skip these records.
int contigId = 0;
seqan::CharString const & contigName = nameStore(bamStream.bamIOContext)[record.rId];
if (!getIdByName(faiIndex, contigName, contigId))
{
std::cerr << "ERROR: Alignment to unknown contig " << contigId << "!\n";
return 1;
}
unsigned binNo = record.pos / options.windowSize;
bins[contigId][binNo].coverage += 1;
}
std::cerr << "DONE\n";
// -----------------------------------------------------------------------
// Write Output
// -----------------------------------------------------------------------
std::ostream * out = &std::cout;
std::ofstream outFile;
if (options.outPath != "-")
{
outFile.open(toCString(options.outPath), std::ios::binary | std::ios::out);
if (!outFile.good())
{
std::cerr << "ERROR: Could not open output file " << options.outPath << "!\n";
return 1;
}
out = &outFile;
}
(*out) << "#BIN\tREF_NAME\tREF_BIN\tBIN_BEGIN\tBIN_LENGTH\tCOVERAGE\tCG_CONTENT\n";
for (unsigned i = 0, globalBin = 0; i < length(bins); ++i)
{
for (unsigned refBin = 0; refBin < length(bins[i]); ++refBin, ++globalBin)
{
(*out) << globalBin << '\t'
<< sequenceName(faiIndex, i) << '\t'
<< refBin << '\t'
<< refBin * options.windowSize << '\t'
<< bins[i][refBin].length << '\t'
<< bins[i][refBin].coverage << '\t'
<< bins[i][refBin].cgContent << '\n';
}
}
if (options.verbosity >= 2)
std::cerr << "Took " << (sysTime() - startTime) << " s\n";
return 0;
}
bool VcfMaterializer::_materializeNext(seqan::Dna5String & seq,
MethylationLevels * levels,
std::vector<SmallVarInfo> & varInfos,
std::vector<std::pair<int, int> > & breakpoints,
int & rID,
int & haplotype)
{
if (levels)
SEQAN_CHECK(!empty(methFastaFileName), "Must initialize with methylation FASTA file for levels");
if (empty(vcfFileName))
{
if (currRID >= (int)(numSeqs(faiIndex) - 1))
return false;
currRID += 1;
rID = currRID;
readSequence(seq, faiIndex, currRID);
if (levels && !empty(methFastaFileName))
{
_loadLevels(currRID);
swap(*levels, currentLevels);
}
// Build identity PositionMap.
TJournalEntries journal;
reinit(journal, length(seq));
posMap.reinit(journal);
GenomicInterval gi(0, length(seq), 0, length(seq));
seqan::String<PositionMap::TInterval> intervals;
appendValue(intervals, PositionMap::TInterval(gi.svBeginPos, gi.svEndPos, gi));
createIntervalTree(posMap.svIntervalTree, intervals);
createIntervalTree(posMap.svIntervalTreeSTL, intervals);
return true;
}
// Number of sequences.
int numSeqs = length(contigNames(context(vcfFileIn)));
// Stop if there are no more haplotypes to materialize.
if (currRID >= (numSeqs - 1) && nextHaplotype == numHaplotypes)
return false;
// Load variants for next contig if necessary.
if (currRID == -1 || nextHaplotype == numHaplotypes)
{
currRID += 1;
nextHaplotype = 0;
_loadVariantsForContig(contigVariants, currRID);
readSequence(contigSeq, faiIndex, currRID);
if (levels && !empty(methFastaFileName))
_loadLevels(currRID);
}
// Materialize variants for the current haplotype.
VariantMaterializer varMat(rng, contigVariants, *methOptions);
if (levels)
varMat.run(seq, posMap, *levels, varInfos, breakpoints, contigSeq, currentLevels, nextHaplotype);
else
varMat.run(seq, posMap, varInfos, breakpoints, contigSeq, nextHaplotype);
// Write out rID and haploty
rID = currRID;
haplotype = nextHaplotype++;
return true;
}