int ReadReference::execute(int argc, char **argv)
{
static const int UNSPECIFIED_INT = -1;
String refFile = "";
String refName = "";
int start = UNSPECIFIED_INT;
int numBases = UNSPECIFIED_INT;
int end = UNSPECIFIED_INT;
bool params = false;
// Read in the parameters.
ParameterList inputParameters;
BEGIN_LONG_PARAMETERS(longParameterList)
LONG_STRINGPARAMETER("refFile", &refFile)
LONG_STRINGPARAMETER("refName", &refName)
LONG_INTPARAMETER("start", &start)
LONG_INTPARAMETER("end", &end)
LONG_INTPARAMETER("numBases", &numBases)
LONG_PARAMETER("params", ¶ms)
LONG_PHONEHOME(VERSION)
END_LONG_PARAMETERS();
inputParameters.Add(new LongParameters ("Input Parameters",
longParameterList));
// parameters start at index 2 rather than 1.
inputParameters.Read(argc, argv, 2);
if((refName == "") || (start == UNSPECIFIED_INT) ||
((end == UNSPECIFIED_INT) && (numBases == UNSPECIFIED_INT)))
{
usage();
inputParameters.Status();
std::cerr << "Missing Required Parameter\n\n";
return(-1);
}
if((end != UNSPECIFIED_INT) && (numBases != UNSPECIFIED_INT))
{
usage();
inputParameters.Status();
std::cerr << "Only --end or --numBases can be specified\n\n";
return(-1);
}
else if(numBases != UNSPECIFIED_INT)
{
end = start + numBases;
}
if(params)
{
inputParameters.Status();
}
// Open the reference.
GenomeSequence reference(refFile);
uint32_t refStart =
reference.getGenomePosition(refName.c_str());
if(refStart == INVALID_GENOME_INDEX)
{
std::cerr << "Reference Name: " << refName.c_str()
<< " not found in the reference file\n";
return(-1);
}
std::string refString;
reference.getString(refString, refStart + start, end - start);
std::cout << refString << std::endl;
return(0);
}
int Revert::execute(int argc, char **argv)
{
// Extract command line arguments.
String inFile = "";
String outFile = "";
bool cigar = false;
bool qual = false;
bool noeof = false;
bool params = false;
bool rmBQ = false;
String rmTags = "";
myKeepTags = false;
ParameterList inputParameters;
BEGIN_LONG_PARAMETERS(longParameterList)
LONG_STRINGPARAMETER("in", &inFile)
LONG_STRINGPARAMETER("out", &outFile)
LONG_PARAMETER("cigar", &cigar)
LONG_PARAMETER("qual", &qual)
LONG_PARAMETER("keepTags", &myKeepTags)
LONG_PARAMETER("rmBQ", &rmBQ)
LONG_STRINGPARAMETER("rmTags", &rmTags)
LONG_PARAMETER("noeof", &noeof)
LONG_PARAMETER("params", ¶ms)
LONG_PHONEHOME(VERSION)
END_LONG_PARAMETERS();
inputParameters.Add(new LongParameters ("Input Parameters",
longParameterList));
// parameters start at index 2 rather than 1.
inputParameters.Read(argc, argv, 2);
// If no eof block is required for a bgzf file, set the bgzf file type to
// not look for it.
if(noeof)
{
// Set that the eof block is not required.
BgzfFileType::setRequireEofBlock(false);
}
// Check to see if the in file was specified, if not, report an error.
if(inFile == "")
{
usage();
inputParameters.Status();
// In file was not specified but it is mandatory.
std::cerr << "--in is a mandatory argument, "
<< "but was not specified" << std::endl;
return(-1);
}
if(outFile == "")
{
usage();
inputParameters.Status();
// In file was not specified but it is mandatory.
std::cerr << "--out is a mandatory argument, "
<< "but was not specified" << std::endl;
return(-1);
}
if(params)
{
inputParameters.Status();
}
// Open the input file for reading.
SamFile samIn;
samIn.OpenForRead(inFile);
// Open the output file for writing.
SamFile samOut;
samOut.OpenForWrite(outFile);
// Read the sam header.
SamFileHeader samHeader;
samIn.ReadHeader(samHeader);
// Write the sam header.
samOut.WriteHeader(samHeader);
SamRecord samRecord;
// Set returnStatus to success. It will be changed to the
// failure reason if any of the writes or updates fail.
SamStatus::Status returnStatus = SamStatus::SUCCESS;
// Keep reading records until ReadRecord returns false.
while(samIn.ReadRecord(samHeader, samRecord))
{
// Update the cigar & position.
if(cigar)
{
if(!updateCigar(samRecord))
{
// Failed to update the cigar & position.
fprintf(stderr, "%s\n", samIn.GetStatusMessage());
returnStatus = samIn.GetStatus();
}
}
if(qual)
{
if(!updateQual(samRecord))
{
// Failed to update the quality.
fprintf(stderr, "%s\n", samIn.GetStatusMessage());
returnStatus = samIn.GetStatus();
}
}
if(rmBQ)
{
if(!removeBQ(samRecord))
{
// Failed to remove BQ.
fprintf(stderr, "%s\n", samIn.GetStatusMessage());
returnStatus = samIn.GetStatus();
}
}
if(rmTags != "")
{
if(!samRecord.rmTags(rmTags.c_str()))
{
// Failed to remove the specified tags.
fprintf(stderr, "%s\n", samIn.GetStatusMessage());
returnStatus = samIn.GetStatus();
}
}
// Successfully read a record from the file, so write it.
if(!samOut.WriteRecord(samHeader, samRecord))
{
// Failed to write a record.
fprintf(stderr, "%s\n", samOut.GetStatusMessage());
returnStatus = samOut.GetStatus();
}
}
std::cerr << std::endl << "Number of records read = " <<
samIn.GetCurrentRecordCount() << std::endl;
std::cerr << "Number of records written = " <<
samOut.GetCurrentRecordCount() << std::endl;
// Since the reads were successful, return the status based
// on the status of the writes. If any failed, return
// their failure status.
return(returnStatus);
}
int Convert::execute(int argc, char **argv)
{
// Extract command line arguments.
String inFile = "";
String outFile = "";
String refFile = "";
bool lshift = false;
bool noeof = false;
bool params = false;
bool useBases = false;
bool useEquals = false;
bool useOrigSeq = false;
bool recover = false;
ParameterList inputParameters;
BEGIN_LONG_PARAMETERS(longParameterList)
LONG_STRINGPARAMETER("in", &inFile)
LONG_STRINGPARAMETER("out", &outFile)
LONG_STRINGPARAMETER("refFile", &refFile)
LONG_PARAMETER("lshift", &lshift)
LONG_PARAMETER("noeof", &noeof)
LONG_PARAMETER("recover", &recover)
LONG_PARAMETER("params", ¶ms)
LONG_PARAMETER_GROUP("SequenceConversion")
EXCLUSIVE_PARAMETER("useBases", &useBases)
EXCLUSIVE_PARAMETER("useEquals", &useEquals)
EXCLUSIVE_PARAMETER("useOrigSeq", &useOrigSeq)
LONG_PHONEHOME(VERSION)
END_LONG_PARAMETERS();
inputParameters.Add(new LongParameters ("Input Parameters",
longParameterList));
// parameters start at index 2 rather than 1.
inputParameters.Read(argc, argv, 2);
// If no eof block is required for a bgzf file, set the bgzf file type to
// not look for it.
if(noeof)
{
// Set that the eof block is not required.
BgzfFileType::setRequireEofBlock(false);
}
// Check to see if the in file was specified, if not, report an error.
if(inFile == "")
{
printUsage(std::cerr);
inputParameters.Status();
// In file was not specified but it is mandatory.
std::cerr << "--in is a mandatory argument, "
<< "but was not specified" << std::endl;
return(-1);
}
if(outFile == "")
{
printUsage(std::cerr);
inputParameters.Status();
// In file was not specified but it is mandatory.
std::cerr << "--out is a mandatory argument, "
<< "but was not specified" << std::endl;
return(-1);
}
// Check to see if the ref file was specified.
// Open the reference.
GenomeSequence* refPtr = NULL;
if(refFile != "")
{
refPtr = new GenomeSequence(refFile);
}
SamRecord::SequenceTranslation translation;
if((useBases) && (refPtr != NULL))
{
translation = SamRecord::BASES;
}
else if((useEquals) && (refPtr != NULL))
{
translation = SamRecord::EQUAL;
}
else
{
useOrigSeq = true;
translation = SamRecord::NONE;
}
if(params)
{
inputParameters.Status();
}
// Open the input file for reading.
SamFile samIn;
if(recover) samIn.setAttemptRecovery(true);
samIn.OpenForRead(inFile);
// Open the output file for writing.
SamFile samOut;
samOut.OpenForWrite(outFile);
samOut.SetWriteSequenceTranslation(translation);
samOut.SetReference(refPtr);
// Read the sam header.
SamFileHeader samHeader;
samIn.ReadHeader(samHeader);
// Write the sam header.
samOut.WriteHeader(samHeader);
SamRecord samRecord;
// Set returnStatus to success. It will be changed
// to the failure reason if any of the writes fail.
SamStatus::Status returnStatus = SamStatus::SUCCESS;
while(1) {
try {
// Keep reading records until ReadRecord returns false.
while(samIn.ReadRecord(samHeader, samRecord))
{
// left shift if necessary.
if(lshift)
{
samRecord.shiftIndelsLeft();
}
// Successfully read a record from the file, so write it.
if(!samOut.WriteRecord(samHeader, samRecord))
{
// Failed to write a record.
fprintf(stderr, "%s\n", samOut.GetStatusMessage());
returnStatus = samOut.GetStatus();
}
}
break;
} catch (std::runtime_error e) {
std::cerr << "Caught runtime error: " << e.what() << "\n";
if(!recover) {
std::cerr << "Corrupted BAM file detected - consider using --recover option.\n";
break;
}
std::cerr << "Attempting to resync at next good BGZF block and BAM record.\n";
// XXX need to resync SamFile stream here
bool rc = samIn.attemptRecoverySync(checkSignature, SIGNATURE_LENGTH);
if(rc) {
std::cerr << "Successful resync - some data lost.\n";
continue; // succeeded
}
std::cerr << "Failed to re-sync on data stream.\n";
break; // failed to resync
}
}
std::cerr << std::endl << "Number of records read = " <<
samIn.GetCurrentRecordCount() << std::endl;
std::cerr << "Number of records written = " <<
samOut.GetCurrentRecordCount() << std::endl;
if(refPtr != NULL)
{
delete(refPtr);
}
// Since the reads were successful, return the status based
// on the status of the writes. If any failed, return
// their failure status.
return(returnStatus);
}
int VcfConvert::execute(int argc, char **argv)
{
String refFile = "";
String inputVcf = "";
String outputVcf = "";
String refName = "";
bool uncompress = false;
bool params = false;
bool noeof = false;
// Read in the parameters.
ParameterList inputParameters;
BEGIN_LONG_PARAMETERS(longParameterList)
LONG_PARAMETER_GROUP("Required Parameters")
LONG_STRINGPARAMETER("in", &inputVcf)
LONG_STRINGPARAMETER("out", &outputVcf)
LONG_PARAMETER_GROUP("Optional Parameters")
LONG_PARAMETER("uncompress", &uncompress)
LONG_STRINGPARAMETER("refName", &refName)
LONG_PARAMETER("noeof", &noeof)
LONG_PARAMETER("params", ¶ms)
LONG_PHONEHOME(VERSION)
END_LONG_PARAMETERS();
inputParameters.Add(new LongParameters ("Input Parameters",
longParameterList));
inputParameters.Read(argc-1, &(argv[1]));
// Check that all files were specified.
if(inputVcf == "")
{
usage();
inputParameters.Status();
std::cerr << "Missing \"--in\", a required parameter.\n\n";
return(-1);
}
if(outputVcf == "")
{
usage();
inputParameters.Status();
std::cerr << "Missing \"--out\", a required parameter.\n\n";
return(-1);
}
if(params)
{
inputParameters.Status();
}
// If no eof block is required for a bgzf file, set the bgzf file type to
// not look for it.
if(noeof)
{
// Set that the eof block is not required.
BgzfFileType::setRequireEofBlock(false);
}
VcfFileReader inFile;
VcfFileWriter outFile;
VcfHeader header;
// Open the file.
inFile.open(inputVcf, header);
if(refName != "")
{
inFile.setReadSection(refName.c_str());
}
if(uncompress)
{
outFile.open(outputVcf, header, InputFile::DEFAULT);
}
else
{
outFile.open(outputVcf, header);
}
VcfRecord record;
int numRecords = 0;
while(inFile.readRecord(record))
{
++numRecords;
outFile.writeRecord(record);
}
inFile.close();
std::cerr << "NumRecords: " << numRecords << "\n";
return(0);
}
int ClipOverlap::execute(int argc, char **argv)
{
// Extract command line arguments.
String inFile = "";
String outFile = "";
String storeOrig = "";
bool readName = false;
bool noRNValidate = false;
bool stats = false;
int poolSize = DEFAULT_POOL_SIZE;
bool unmapped = false;
bool noeof = false;
bool params = false;
String excludeFlags = "0xF0C";
// TODO, cleanup legacy parameters
ParameterList inputParameters;
BEGIN_LONG_PARAMETERS(longParameterList)
LONG_PARAMETER_GROUP("Required Parameters")
LONG_STRINGPARAMETER("in", &inFile)
LONG_STRINGPARAMETER("out", &outFile)
LONG_PARAMETER_GROUP("Optional Parameters")
LONG_STRINGPARAMETER("storeOrig", &storeOrig)
LONG_PARAMETER("readName", &readName)
LONG_PARAMETER ("noRNValidate", &noRNValidate)
LONG_PARAMETER ("stats", &stats)
LONG_PARAMETER ("overlapsOnly", &myOverlapsOnly)
LONG_STRINGPARAMETER ("excludeFlags", &excludeFlags)
LONG_PARAMETER("unmapped", &unmapped)
LONG_PARAMETER("noeof", &noeof)
LONG_PARAMETER("params", ¶ms)
LONG_PARAMETER_GROUP("Coordinate Processing Optional Parameters")
LONG_INTPARAMETER("poolSize", &poolSize)
LONG_PARAMETER("poolSkipOverlap", &myPoolSkipOverlap)
LONG_PHONEHOME(VERSION)
BEGIN_LEGACY_PARAMETERS()
LONG_PARAMETER ("clipsOnly", &myOverlapsOnly)
LONG_PARAMETER("poolSkipClip", &myPoolSkipOverlap)
END_LONG_PARAMETERS();
inputParameters.Add(new LongParameters ("Input Parameters",
longParameterList));
// parameters start at index 2 rather than 1.
inputParameters.Read(argc, argv, 2);
// If no eof block is required for a bgzf file, set the bgzf file type to
// not look for it.
if(noeof)
{
// Set that the eof block is not required.
BgzfFileType::setRequireEofBlock(false);
}
// Check to see if the in file was specified, if not, report an error.
if(inFile == "")
{
printUsage(std::cerr);
inputParameters.Status();
// In file was not specified but it is mandatory.
std::cerr << "--in is a mandatory argument, "
<< "but was not specified" << std::endl;
return(-1);
}
// Check to see if the out file was specified, if not, report an error.
if(outFile == "")
{
printUsage(std::cerr);
inputParameters.Status();
// Out file was not specified but it is mandatory.
std::cerr << "--out is a mandatory argument, "
<< "but was not specified" << std::endl;
return(-1);
}
if((storeOrig.Length() != 0) && (storeOrig.Length() != 2))
{
printUsage(std::cerr);
inputParameters.Status();
std::cerr << "--storeOrig tag name must be 2 characters.\n";
return(-1);
}
myOverlapHandler = new OverlapClipLowerBaseQual();
if(myOverlapHandler == NULL)
{
printUsage(std::cerr);
inputParameters.Status();
std::cerr << "Failed to allocate the overlap handler\n";
return(-1);
}
if(unmapped)
{
myOverlapHandler->markAsUnmapped();
}
// Setup the overlap handler.
myOverlapHandler->keepStats(stats);
if(storeOrig.Length() != 0)
{
myOverlapHandler->storeOrigCigar(storeOrig);
}
myIntExcludeFlags = excludeFlags.AsInteger();
if(params)
{
inputParameters.Status();
}
// For each step process the file.
// Open the files & read/write the sam header.
SamStatus::Status runStatus = SamStatus::SUCCESS;
for(int i = 1; i <= myOverlapHandler->numSteps(); i++)
{
// Open the file for reading.
mySamHeader.resetHeader();
SamFile samIn(inFile, SamFile::READ, &mySamHeader);
SamFile* samOutPtr = NULL;
// Check if writing, if so, open the output file.
if(i == myOverlapHandler->numSteps())
{
samOutPtr = new SamFile(outFile, SamFile::WRITE, &mySamHeader);
}
if(readName)
{
if(!noRNValidate)
{
samIn.setSortedValidation(SamFile::QUERY_NAME);
}
runStatus = handleSortedByReadName(samIn, samOutPtr);
}
else
{
// Coordinate sorted, so work with the pools.
samIn.setSortedValidation(SamFile::COORDINATE);
myPool.setMaxAllocatedRecs(poolSize);
// Reset the number of failures
myNumMateFailures = 0;
myNumPoolFail = 0;
myNumPoolFailNoHandle = 0;
myNumPoolFailHandled = 0;
myNumOutOfOrder = 0;
// Run by coordinate
if(samOutPtr != NULL)
{
// Setup the output buffer for writing.
SamCoordOutput outputBuffer(myPool);
outputBuffer.setOutputFile(samOutPtr, &mySamHeader);
runStatus = handleSortedByCoord(samIn, &outputBuffer);
// Cleanup the output buffer.
if(!outputBuffer.flushAll())
{
std::cerr << "ERROR: Failed to flush the output buffer\n";
runStatus = SamStatus::FAIL_IO;
}
}
else
{
runStatus = handleSortedByCoord(samIn, NULL);
}
}
if(runStatus != SamStatus::SUCCESS)
{
break;
}
// Close the input file, it will be reopened if there are
// multiple steps.
samIn.Close();
if(samOutPtr != NULL)
{
samOutPtr->Close();
delete samOutPtr;
samOutPtr = NULL;
}
}
// Done processing.
// Print Stats
myOverlapHandler->printStats();
if(myNumMateFailures != 0)
{
std::cerr << "WARNING: did not find expected overlapping mates for "
<< myNumMateFailures << " records." << std::endl;
}
if(myNumPoolFail != 0)
{
// Had to skip clipping some records due to running out of
// memory and not being able to wait for the mate.
std::cerr << "WARNING: " << myNumPoolFail
<< " record pool failures\n";
if(myNumPoolFailNoHandle != 0)
{
std::cerr << "Due to hitting the max record poolSize, skipped handling "
<< myNumPoolFailNoHandle << " records." << std::endl;
}
if(myNumPoolFailHandled != 0)
{
std::cerr << "Due to hitting the max record poolSize, default handled "
<< myNumPoolFailHandled << " records." << std::endl;
}
if(myNumOutOfOrder != 0)
{
std::cerr << "WARNING: Resulting File out of Order by "
<< myNumOutOfOrder << " records.\n";
}
}
if(runStatus == SamStatus::SUCCESS)
{
if(myNumPoolFail == 0)
{
std::cerr << "Completed ClipOverlap Successfully.\n";
}
else
{
runStatus = SamStatus::NO_MORE_RECS;
std::cerr << "Completed ClipOverlap with WARNINGS.\n";
}
}
else
{
std::cerr << "Failed to complete ClipOverlap.\n";
}
return(runStatus);
}
int Bam2FastQ::execute(int argc, char **argv)
{
// Extract command line arguments.
String inFile = "";
bool readName = false;
String refFile = "";
String firstOut = "";
String secondOut = "";
String unpairedOut = "";
bool interleave = false;
bool noeof = false;
bool gzip = false;
bool params = false;
myOutBase = "";
myNumMateFailures = 0;
myNumPairs = 0;
myNumUnpaired = 0;
mySplitRG = false;
myQField = "";
myNumQualTagErrors = 0;
myReverseComp = true;
myRNPlus = false;
myFirstRNExt = DEFAULT_FIRST_EXT;
mySecondRNExt = DEFAULT_SECOND_EXT;
myCompression = InputFile::DEFAULT;
ParameterList inputParameters;
BEGIN_LONG_PARAMETERS(longParameterList)
LONG_PARAMETER_GROUP("Required Parameters")
LONG_STRINGPARAMETER("in", &inFile)
LONG_PARAMETER_GROUP("Optional Parameters")
LONG_PARAMETER("readName", &readName)
LONG_PARAMETER("splitRG", &mySplitRG)
LONG_STRINGPARAMETER("qualField", &myQField)
LONG_PARAMETER("merge", &interleave)
LONG_STRINGPARAMETER("refFile", &refFile)
LONG_STRINGPARAMETER("firstRNExt", &myFirstRNExt)
LONG_STRINGPARAMETER("secondRNExt", &mySecondRNExt)
LONG_PARAMETER("rnPlus", &myRNPlus)
LONG_PARAMETER("noReverseComp", &myReverseComp)
LONG_PARAMETER("gzip", &gzip)
LONG_PARAMETER("noeof", &noeof)
LONG_PARAMETER("params", ¶ms)
LONG_PARAMETER_GROUP("Optional OutputFile Names")
LONG_STRINGPARAMETER("outBase", &myOutBase)
LONG_STRINGPARAMETER("firstOut", &firstOut)
LONG_STRINGPARAMETER("secondOut", &secondOut)
LONG_STRINGPARAMETER("unpairedOut", &unpairedOut)
LONG_PHONEHOME(VERSION)
END_LONG_PARAMETERS();
inputParameters.Add(new LongParameters ("Input Parameters",
longParameterList));
// parameters start at index 2 rather than 1.
inputParameters.Read(argc, argv, 2);
// If no eof block is required for a bgzf file, set the bgzf file type to
// not look for it.
if(noeof)
{
// Set that the eof block is not required.
BgzfFileType::setRequireEofBlock(false);
}
if(gzip)
{
myCompression = InputFile::GZIP;
}
// Check to see if the in file was specified, if not, report an error.
if(inFile == "")
{
usage();
inputParameters.Status();
// In file was not specified but it is mandatory.
std::cerr << "--in is a mandatory argument, "
<< "but was not specified" << std::endl;
return(-1);
}
// Cannot specify both interleaved & secondOut since secondOut would be N/A.
if(interleave && !secondOut.IsEmpty())
{
usage();
inputParameters.Status();
std::cerr << "ERROR: Cannot specify --merge & --secondOut.\n";
return(-1);
}
// Cannot specify both interleaved & secondOut since secondOut would be N/A.
if(interleave && !secondOut.IsEmpty())
{
usage();
inputParameters.Status();
std::cerr << "ERROR: Cannot specify --merge & --secondOut.\n";
return(-1);
}
// Cannot specify both splitRG & firstOut/secondOut/unpairedOut
// since it needs a different file for each RG.
if(mySplitRG && (!firstOut.IsEmpty() ||
!secondOut.IsEmpty() || !unpairedOut.IsEmpty()))
{
usage();
inputParameters.Status();
std::cerr << "ERROR: Cannot specify --splitRG & --firstOut/--secondOut/--unpairedOut.\n";
std::cerr << "Use --outBase instead.\n";
return(-1);
}
// Cannot specify splitRG & output to stdout.
if(mySplitRG && (myOutBase[0] == '-'))
{
usage();
inputParameters.Status();
std::cerr << "ERROR: Cannot specify --splitRG & write to stdout.\n";
return(-1);
}
// Check to see if the out file was specified, if not, generate it from
// the input filename.
if(myOutBase == "")
{
// Just remove the extension from the input filename.
int extStart = inFile.FastFindLastChar('.');
if(extStart <= 0)
{
myOutBase = inFile;
}
else
{
myOutBase = inFile.Left(extStart);
}
}
if(mySplitRG)
{
std::string fqList = myOutBase.c_str();
fqList += ".list";
myFqList = ifopen(fqList.c_str(), "w");
ifprintf(myFqList, "MERGE_NAME\tFASTQ1\tFASTQ2\tRG\n");
}
// Check to see if the first/second/single-ended were specified and
// if not, set them.
myFirstFileNameExt = "_1.fastq";
mySecondFileNameExt = "_2.fastq";
myUnpairedFileNameExt = ".fastq";
if(interleave)
{
myFirstFileNameExt = "_interleaved.fastq";
myFirstFileNameExt = "_interleaved.fastq";
}
getFileName(firstOut, myFirstFileNameExt);
getFileName(secondOut, mySecondFileNameExt);
getFileName(unpairedOut, myUnpairedFileNameExt);
if(params)
{
inputParameters.Status();
}
// Open the files for reading/writing.
// Open prior to opening the output files,
// so if there is an error, the outputs don't get created.
SamFile samIn;
samIn.OpenForRead(inFile, &mySamHeader);
// Skip non-primary reads.
samIn.SetReadFlags(0, 0x0100);
// Open the output files if not splitting RG
if(!mySplitRG)
{
myUnpairedFile = ifopen(unpairedOut, "w", myCompression);
// Only open the first file if it is different than an already opened file.
if(firstOut != unpairedOut)
{
myFirstFile = ifopen(firstOut, "w", myCompression);
}
else
{
myFirstFile = myUnpairedFile;
}
// If it is interleaved or the 2nd file is not a new name, set it appropriately.
if(interleave || secondOut == firstOut)
{
mySecondFile = myFirstFile;
}
else if(secondOut == unpairedOut)
{
mySecondFile = myUnpairedFile;
}
else
{
mySecondFile = ifopen(secondOut, "w", myCompression);
}
if(myUnpairedFile == NULL)
{
std::cerr << "Failed to open " << unpairedOut
<< " so can't convert bam2FastQ.\n";
return(-1);
}
if(myFirstFile == NULL)
{
std::cerr << "Failed to open " << firstOut
<< " so can't convert bam2FastQ.\n";
return(-1);
}
if(mySecondFile == NULL)
{
std::cerr << "Failed to open " << secondOut
<< " so can't convert bam2FastQ.\n";
return(-1);
}
}
if((readName) || (strcmp(mySamHeader.getSortOrder(), "queryname") == 0))
{
readName = true;
}
else
{
// defaulting to coordinate sorted.
samIn.setSortedValidation(SamFile::COORDINATE);
}
// Setup the '=' translation if the reference was specified.
if(!refFile.IsEmpty())
{
GenomeSequence* refPtr = new GenomeSequence(refFile);
samIn.SetReadSequenceTranslation(SamRecord::BASES);
samIn.SetReference(refPtr);
}
SamRecord* recordPtr;
int16_t samFlag;
SamStatus::Status returnStatus = SamStatus::SUCCESS;
while(returnStatus == SamStatus::SUCCESS)
{
recordPtr = myPool.getRecord();
if(recordPtr == NULL)
{
// Failed to allocate a new record.
throw(std::runtime_error("Failed to allocate a new SAM/BAM record"));
}
if(!samIn.ReadRecord(mySamHeader, *recordPtr))
{
// Failed to read a record.
returnStatus = samIn.GetStatus();
continue;
}
// Have a record. Check to see if it is a pair or unpaired read.
samFlag = recordPtr->getFlag();
if(SamFlag::isPaired(samFlag))
{
if(readName)
{
handlePairedRN(*recordPtr);
}
else
{
handlePairedCoord(*recordPtr);
}
}
else
{
++myNumUnpaired;
writeFastQ(*recordPtr, myUnpairedFile,
myUnpairedFileNameExt);
}
}
// Flush All
cleanUpMateMap(0, true);
if(returnStatus == SamStatus::NO_MORE_RECS)
{
returnStatus = SamStatus::SUCCESS;
}
samIn.Close();
closeFiles();
// Output the results
std::cerr << "\nFound " << myNumPairs << " read pairs.\n";
std::cerr << "Found " << myNumUnpaired << " unpaired reads.\n";
if(myNumMateFailures != 0)
{
std::cerr << "Failed to find mates for " << myNumMateFailures
<< " reads, so they were written as unpaired\n"
<< " (not included in either of the above counts).\n";
}
if(myNumQualTagErrors != 0)
{
std::cerr << myNumQualTagErrors << " records did not have tag "
<< myQField.c_str() << " or it was invalid, so the quality field was used for those records.\n";
}
return(returnStatus);
}
int VcfSplit::execute(int argc, char **argv)
{
String refFile = "";
String inputVcf = "";
String outputVcfBase = "";
String refName = "";
bool uncompress = false;
bool params = false;
bool noeof = false;
// Read in the parameters.
ParameterList inputParameters;
BEGIN_LONG_PARAMETERS(longParameterList)
LONG_PARAMETER_GROUP("Required Parameters")
LONG_STRINGPARAMETER("in", &inputVcf)
LONG_STRINGPARAMETER("obase", &outputVcfBase)
LONG_PARAMETER_GROUP("Optional Parameters")
LONG_PARAMETER("uncompress", &uncompress)
LONG_STRINGPARAMETER("refName", &refName)
LONG_PARAMETER("noeof", &noeof)
LONG_PARAMETER("params", ¶ms)
LONG_PHONEHOME(VERSION)
END_LONG_PARAMETERS();
inputParameters.Add(new LongParameters ("Input Parameters",
longParameterList));
inputParameters.Read(argc-1, &(argv[1]));
// Check that all files were specified.
if(inputVcf == "")
{
usage();
inputParameters.Status();
std::cerr << "Missing \"--in\", a required parameter.\n\n";
return(-1);
}
if(outputVcfBase == "")
{
usage();
inputParameters.Status();
std::cerr << "Missing \"--obase\", a required parameter.\n\n";
return(-1);
}
outputVcfBase += ".";
if(params)
{
inputParameters.Status();
}
// If no eof block is required for a bgzf file, set the bgzf file type to
// not look for it.
if(noeof)
{
// Set that the eof block is not required.
BgzfFileType::setRequireEofBlock(false);
}
VcfFileReader inFile;
std::map<std::string, VcfFileWriter*> outFiles;
VcfHeader header;
// Open the file.
inFile.open(inputVcf, header);
if(refName != "")
{
inFile.setReadSection(refName.c_str());
}
VcfRecord record;
int numRecords = 0;
std::string prevChr = "";
std::string chr = "";
VcfFileWriter* outFilePtr = 0;
std::string outName = "";
while(inFile.readRecord(record))
{
++numRecords;
chr = record.getChromStr();
if((outFilePtr == 0) || (chr != prevChr))
{
outFilePtr = outFiles[chr];
if(outFilePtr == 0)
{
outFilePtr = new VcfFileWriter();
outFiles[chr] = outFilePtr;
outName = outputVcfBase.c_str();
if(chr.substr(0,3) != "chr")
{
outName += "chr";
}
outName += chr + ".vcf";
// chr not in outFile list.
if(uncompress)
{
outFilePtr->open(outName.c_str(), header, InputFile::DEFAULT);
}
else
{
outName += ".gz";
outFilePtr->open(outName.c_str(), header);
}
}
}
outFilePtr->writeRecord(record);
}
inFile.close();
for (std::map<std::string,VcfFileWriter*>::iterator it = outFiles.begin();
it != outFiles.end(); ++it)
{
if(it->second != 0)
{
it->second->close();
it->second = 0;
}
}
std::cerr << "NumRecords: " << numRecords << "\n";
return(0);
}
int VcfConsensus::execute(int argc, char ** argv)
{
String vcfName1;
String vcfName2;
String vcfName3;
String outputFileName;
bool uncompress = false;
bool params = false;
// Read in the parameters.
ParameterList inputParameters;
BEGIN_LONG_PARAMETERS(longParameterList)
LONG_PARAMETER_GROUP("Required Parameters")
LONG_STRINGPARAMETER("in1", &vcfName1)
LONG_STRINGPARAMETER("in2", &vcfName2)
LONG_STRINGPARAMETER("in3", &vcfName3)
LONG_STRINGPARAMETER("out", &outputFileName)
LONG_PARAMETER_GROUP("Optional Parameters")
LONG_PARAMETER("uncompress", &uncompress)
LONG_PARAMETER("params", ¶ms)
LONG_PHONEHOME(VERSION)
END_LONG_PARAMETERS();
inputParameters.Add(new LongParameters ("Input Parameters",
longParameterList));
inputParameters.Read(argc-1, &(argv[1]));
std::string gtField = "GT";
VcfFileReader vcf1;
VcfFileReader vcf2;
VcfFileReader vcf3;
VcfHeader header1;
VcfHeader header2;
VcfHeader header3;
VcfRecord record1;
VcfRecord record2;
VcfRecord record3;
VcfRecordGenotype* genotypeInfoPtr1 = NULL;
VcfRecordGenotype* genotypeInfoPtr2 = NULL;
VcfRecordGenotype* genotypeInfoPtr3 = NULL;
unsigned int numMissing2 = 0;
unsigned int numMissing3 = 0;
unsigned int numMismatchRefAlt = 0;
unsigned int numMissingGT1 = 0;
const unsigned int myMaxErrors = 4;
// Check that the required parameters were set.
if(vcfName1 == "")
{
usage();
inputParameters.Status();
std::cerr << "Missing \"--in1\", a required parameter.\n\n";
return(-1);
}
if(vcfName2 == "")
{
usage();
inputParameters.Status();
std::cerr << "Missing \"--in2\", a required parameter.\n\n";
return(-1);
}
if(vcfName3 == "")
{
usage();
inputParameters.Status();
std::cerr << "Missing \"--in3\", a required parameter.\n\n";
return(-1);
}
if(outputFileName == "")
{
usage();
inputParameters.Status();
std::cerr << "Missing \"--out\", a required parameter.\n\n";
return(-1);
}
if(params)
{
inputParameters.Status();
}
// Open the files.
vcf1.open(vcfName1, header1);
vcf2.open(vcfName2, header2);
vcf3.open(vcfName3, header3);
// Setup the sample name maps.
int numSamples = header1.getNumSamples();
std::vector<int> sample2Indices;
std::vector<int> sample3Indices;
std::vector<int> removeIndices;
int numSamplesSkipped1 = 0;
int numSamplesSkipped2 = 0;
int numSamplesSkipped3 = 0;
for(int i = 0; i < numSamples; i++)
{
int sm2Index = header2.getSampleIndex(header1.getSampleName(i));
int sm3Index = header3.getSampleIndex(header1.getSampleName(i));
// Look for this sample name in vcf2.
if((sm2Index != -1) && (sm3Index != -1))
{
sample2Indices.push_back(sm2Index);
sample3Indices.push_back(sm3Index);
}
else
{
// Sample not found in all three vcfs.
removeIndices.push_back(i);
++numSamplesSkipped1;
}
}
// Remove samples not found in all 3 vcfs from header1.
// Remove them in reverse order so they are removed from the end of the header first.
VcfSubsetSamples subset1;
subset1.init(header1, true);
for(int i = (removeIndices.size() - 1); i >= 0; i--)
{
subset1.addExcludeSample(header1.getSampleName(removeIndices[i]));
header1.removeSample(removeIndices[i]);
}
// Set numSamples to the new number of samples in header1.
numSamples = header1.getNumSamples();
// Calculate the number of samples skipped for files 2 & 3.
numSamplesSkipped2 = header2.getNumSamples() - sample2Indices.size();
numSamplesSkipped3 = header3.getNumSamples() - sample3Indices.size();
if(numSamplesSkipped1 > 0)
{
std::cerr << "Skipping " << numSamplesSkipped1 << " samples from --in1\n";
}
if(numSamplesSkipped2 > 0)
{
std::cerr << "Skipping " << numSamplesSkipped2 << " samples from --in2\n";
}
if(numSamplesSkipped3 > 0)
{
std::cerr << "Skipping " << numSamplesSkipped3 << " samples from --in3\n";
}
VcfFileWriter outputVcf;
// Open and write the header
if(uncompress)
{
outputVcf.open(outputFileName, header1, InputFile::DEFAULT);
}
else
{
outputVcf.open(outputFileName, header1);
}
const char* chrom1 = NULL;
int pos1 = UNSET_POS;
// Read the first record from vcf2 & vcf3.
vcf2.readRecord(record2);
vcf3.readRecord(record3);
bool newChrom = true;
static std::string prevChrom = "";
uint64_t numAllMatch = 0;
uint64_t num1Match2Only = 0;
uint64_t num1Match3Only = 0;
uint64_t num2Match3Only = 0;
uint64_t numNoMatches = 0;
uint64_t numAllMatch00 = 0;
uint64_t num1Match2Only00 = 0;
uint64_t num1Match3Only00 = 0;
uint64_t num2Match3Only00 = 0;
uint64_t numAllMatch01 = 0;
uint64_t num1Match2Only01 = 0;
uint64_t num1Match3Only01 = 0;
uint64_t num2Match3Only01 = 0;
uint64_t numAllMatch11 = 0;
uint64_t num1Match2Only11 = 0;
uint64_t num1Match3Only11 = 0;
uint64_t num2Match3Only11 = 0;
// Loop through vcf1.
while(vcf1.readRecord(record1, &subset1))
{
chrom1 = record1.getChromStr();
pos1 = record1.get1BasedPosition();
if(strcmp(chrom1, prevChrom.c_str()) == 0)
{
newChrom = false;
}
else
{
prevChrom = chrom1;
newChrom = true;
}
bool found = true;
if(!findPos(newChrom, chrom1, pos1, record2, vcf2))
{
// Failed to find the position, continue to the next position
if(++numMissing2 <= myMaxErrors)
{
std::cerr << "Failed to find " << chrom1 << ":" << pos1
<< " in " << vcfName2 << ", so skipping this pos\n";
}
found = false;
}
if(!findPos(newChrom, chrom1, pos1, record3, vcf3))
{
// Failed to find the position, continue to the next position
if(++numMissing3 <= myMaxErrors)
{
std::cerr << "Failed to find " << chrom1 << ":" << pos1
<< " in " << vcfName3 << ", so skipping this pos\n";
}
found = false;
}
if(found == false)
{
continue;
}
// Found the position in all files.
// Validate that the reference & alternate alleles are the same.
const char* ref1 = record1.getRefStr();
const char* alt1 = record1.getAltStr();
if((strcmp(ref1, record2.getRefStr()) != 0) ||
(strcmp(ref1, record3.getRefStr()) != 0) ||
(strcmp(alt1, record2.getAltStr()) != 0) ||
(strcmp(alt1, record3.getAltStr()) != 0))
{
if(++numMismatchRefAlt <= myMaxErrors)
{
std::cerr << "Mismatching ref/alt found at " << chrom1 << ":" << pos1 << ", so skipping this pos\n";
}
continue;
}
// Get the genotype information for each.
genotypeInfoPtr1 = &(record1.getGenotypeInfo());
genotypeInfoPtr2 = &(record2.getGenotypeInfo());
genotypeInfoPtr3 = &(record3.getGenotypeInfo());
// Loop through all the samples in vcf1.
// Get the Genotype Information.
for(int i = 0; i < numSamples; i++)
{
const std::string* genotypeVal1 = genotypeInfoPtr1->getString(gtField, i);
const std::string* genotypeVal2 = genotypeInfoPtr2->getString(gtField, sample2Indices[i]);
const std::string* genotypeVal3 = genotypeInfoPtr3->getString(gtField, sample3Indices[i]);
// Need to make sure the field was found.
if(genotypeVal1 == NULL)
{
// GT not found in the first record, so just continue.
if(++numMissingGT1 <= myMaxErrors)
{
std::cerr << "Missing GT for " << header1.getSampleName(i) << " in " << vcfName1 << "\n";
}
continue;
}
if(isSame(genotypeVal1, genotypeVal2))
{
// genotypeVal1 is majority, so make no change.
if(isSame(genotypeVal1, genotypeVal3))
{
++numAllMatch;
if(*genotypeVal1 == "0/0")
{
++numAllMatch00;
}
else if((*genotypeVal1 == "0/1") ||
(*genotypeVal1 == "1/0"))
{
++numAllMatch01;
}
if(*genotypeVal1 == "1/1")
{
++numAllMatch11;
}
}
else
{
++num1Match2Only;
if(*genotypeVal1 == "0/0")
{
++num1Match2Only00;
}
else if((*genotypeVal1 == "0/1") ||
(*genotypeVal1 == "1/0"))
{
++num1Match2Only01;
}
if(*genotypeVal1 == "1/1")
{
++num1Match2Only11;
}
}
}
else if(isSame(genotypeVal1, genotypeVal3))
{
// genotypeVal1 is majority, so make no change.
++num1Match3Only;
if(*genotypeVal1 == "0/0")
{
++num1Match3Only00;
}
else if((*genotypeVal1 == "0/1") ||
(*genotypeVal1 == "1/0"))
{
++num1Match3Only01;
}
if(*genotypeVal1 == "1/1")
{
++num1Match3Only11;
}
}
else if(isSame(genotypeVal2, genotypeVal3))
{
// genotypeVal2 is majority, so change genotypeVal1.
genotypeInfoPtr1->setString(gtField, i, *genotypeVal2);
++num2Match3Only;
if(*genotypeVal2 == "0/0")
{
++num2Match3Only00;
}
else if((*genotypeVal2 == "0/1") ||
(*genotypeVal2 == "1/0"))
{
++num2Match3Only01;
}
if(*genotypeVal2 == "1/1")
{
++num2Match3Only11;
}
}
else
{
// None are the same so set to "./."
genotypeInfoPtr1->setString(gtField, i, "./.");
++numNoMatches;
}
} // loop back to vcf1 samples.
// Write this record.
outputVcf.writeRecord(record1);
} // loop back to next vcf1 record.
std::cerr << "\n";
if(numMissing2 > myMaxErrors)
{
std::cerr << "Suppressed "
<< numMissing2 - myMaxErrors
<< " errors about skipped positions because they were not in "
<< vcfName2
<< "\n";
}
if(numMissing3 > myMaxErrors)
{
std::cerr << "Suppressed "
<< numMissing3 - myMaxErrors
<< " errors about skipped positions because they were not in "
<< vcfName3
<< "\n";
}
if(numMismatchRefAlt > myMaxErrors)
{
std::cerr << "Suppressed "
<< numMismatchRefAlt - myMaxErrors
<< " errors about mismatched ref/alt positions\n";
}
if(numMissingGT1 > myMaxErrors)
{
std::cerr << "Suppressed "
<< numMissingGT1 - myMaxErrors
<< " errors about missing GT for "
<< vcfName1
<< "\n";
}
std::cerr << "\n";
// Output the stats.
std::cerr << "File1 = " << vcfName1 << std::endl;
std::cerr << "File2 = " << vcfName2 << std::endl;
std::cerr << "File3 = " << vcfName3 << std::endl;
std::cerr << "\nType\tTotal\t0/0\t0/1|1/0\t1/1\n";
std::cerr << "AllMatched"
<< "\t" << numAllMatch
<< "\t" << numAllMatch00
<< "\t" << numAllMatch01
<< "\t" << numAllMatch11 << std::endl;
std::cerr << "1matched2"
<< "\t" << num1Match2Only
<< "\t" << num1Match2Only00
<< "\t" << num1Match2Only01
<< "\t" << num1Match2Only11 << std::endl;
std::cerr << "1matched3"
<< "\t" << num1Match3Only
<< "\t" << num1Match3Only00
<< "\t" << num1Match3Only01
<< "\t" << num1Match3Only11 << std::endl;
std::cerr << "2matched3"
<< "\t" << num2Match3Only
<< "\t" << num2Match3Only00
<< "\t" << num2Match3Only01
<< "\t" << num2Match3Only11 << std::endl;
std::cerr << "NoneMatched\t" << numNoMatches << std::endl;
return(0);
}
// main function of verifyBamID
int execute(int argc, char** argv) {
printf("verifyBamID %s -- verify identity and purity of sequence data\n"
"(c) 2010-2014 Hyun Min Kang, Goo Jun, and Goncalo Abecasis\n\n", VERSION);
VerifyBamIDArgs args;
ParameterList pl;
BEGIN_LONG_PARAMETERS(longParameters)
LONG_PARAMETER_GROUP("Input Files")
LONG_STRINGPARAMETER("vcf",&args.sVcfFile)
LONG_STRINGPARAMETER("bam",&args.sBamFile)
LONG_STRINGPARAMETER("subset",&args.sSubsetInds)
LONG_STRINGPARAMETER("smID",&args.sSMID)
LONG_PARAMETER_GROUP("VCF analysis options")
LONG_DOUBLEPARAMETER("genoError",&args.genoError)
LONG_DOUBLEPARAMETER("minAF",&args.minAF)
LONG_DOUBLEPARAMETER("minCallRate",&args.minCallRate)
LONG_PARAMETER_GROUP("Individuals to compare with chip data")
EXCLUSIVE_PARAMETER("site",&args.bSiteOnly)
EXCLUSIVE_PARAMETER("self",&args.bSelfOnly)
EXCLUSIVE_PARAMETER("best",&args.bFindBest)
LONG_PARAMETER_GROUP("Chip-free optimization options")
EXCLUSIVE_PARAMETER("free-none",&args.bFreeNone)
EXCLUSIVE_PARAMETER("free-mix",&args.bFreeMixOnly)
EXCLUSIVE_PARAMETER("free-refBias",&args.bFreeRefBiasOnly)
EXCLUSIVE_PARAMETER("free-full",&args.bFreeFull)
LONG_PARAMETER_GROUP("With-chip optimization options")
EXCLUSIVE_PARAMETER("chip-none",&args.bChipNone)
EXCLUSIVE_PARAMETER("chip-mix",&args.bChipMixOnly)
EXCLUSIVE_PARAMETER("chip-refBias",&args.bChipRefBiasOnly)
EXCLUSIVE_PARAMETER("chip-full",&args.bChipFull)
LONG_PARAMETER_GROUP("BAM analysis options")
LONG_PARAMETER("ignoreRG",&args.bIgnoreRG)
LONG_PARAMETER("ignoreOverlapPair",&args.bIgnoreOverlapPair)
LONG_PARAMETER("noEOF",&args.bNoEOF)
LONG_PARAMETER("precise",&args.bPrecise)
LONG_INTPARAMETER("minMapQ",&args.minMapQ)
LONG_INTPARAMETER("maxDepth",&args.maxDepth)
LONG_INTPARAMETER("minQ",&args.minQ)
LONG_INTPARAMETER("maxQ",&args.maxQ)
LONG_DOUBLEPARAMETER("grid",&args.grid)
LONG_PARAMETER_GROUP("Modeling Reference Bias")
LONG_DOUBLEPARAMETER("refRef",&args.pRefRef)
LONG_DOUBLEPARAMETER("refHet",&args.pRefHet)
LONG_DOUBLEPARAMETER("refAlt",&args.pRefAlt)
LONG_PARAMETER_GROUP("Output options")
LONG_STRINGPARAMETER("out",&args.sOutFile)
LONG_PARAMETER("verbose",&args.bVerbose)
LONG_PHONEHOME(VERSION)
END_LONG_PARAMETERS();
pl.Add(new LongParameters("Available Options",longParameters));
pl.Read(argc, argv);
pl.Status();
// check the validity of input files
if ( args.sVcfFile.IsEmpty() ) {
error("--vcf [vcf file] required");
}
if ( args.sBamFile.IsEmpty() ) {
error("--bam [bam file] is required");
}
if ( args.sOutFile.IsEmpty() ) {
error("--out [output prefix] is required");
}
Logger::gLogger = new Logger((args.sOutFile + ".log").c_str(), args.bVerbose);
if ( ! ( args.bSiteOnly || args.bSelfOnly || args.bFindBest ) ) {
warning("--self option was autotomatically turned on by default. Specify --best option if you wanted to check across all possible samples in the VCF");
args.bSelfOnly = true;
}
if ( ( args.maxDepth > 20 ) && ( !args.bPrecise ) ) {
warning("--precise option is not turned on at --maxDepth %d : may be prone to precision errors",args.maxDepth);
}
if ( ( args.bChipRefBiasOnly ) && ( !args.bSelfOnly ) ) {
error("--self must be set for --chip-refBias to work. Skipping..");
}
// check timestamp
time_t t;
time(&t);
Logger::gLogger->writeLog("Analysis started on %s",ctime(&t));
// load arguments
VerifyBamID vbid(&args);
// load input VCF and BAM files
Logger::gLogger->writeLog("Opening Input Files");
vbid.loadFiles(args.sBamFile.c_str(), args.sVcfFile.c_str());
// Check which genotype-free method is used
if ( args.bFreeNone ) { // if no genotype-free mode is tested. skip it
// do nothing for genotype-free estimation
Logger::gLogger->writeLog("Skipping chip-free estimation of sample mixture");
}
else if ( args.bFreeMixOnly ) { // only mixture is estimated.
// genotype-free method
Logger::gLogger->writeLog("Performing chip-free estimation of sample mixture at fixed reference bias parameters (%lf, %lf, %lf)",args.pRefRef,args.pRefHet,args.pRefAlt);
// scan across multiple readgroups
for(int rg=-1; rg < vbid.nRGs - (int)args.bIgnoreRG; ++rg) {
VerifyBamID::mixLLK mix(&vbid);
mix.OptimizeLLK(rg);
Logger::gLogger->writeLog("Optimal per-sample fMix = %lf, LLK0 = %lf, LLK1 = %lf\n",mix.fMix,mix.llk0,mix.llk1);
vbid.mixOut.llk0s[rg+1] = mix.llk0;
vbid.mixOut.llk1s[rg+1] = mix.llk1;
vbid.mixOut.fMixs[rg+1] = mix.fMix;
}
//vbid.mixRefHet = 0.5;
//vbid.mixRefAlt = 0.00;
}
else if ( args.bFreeRefBiasOnly ) {
Logger::gLogger->writeLog("Performing chip-free estimation of reference-bias without sample mixture");
for(int rg=-1; rg < vbid.nRGs - (int)args.bIgnoreRG; ++rg) {
VerifyBamID::refBiasMixLLKFunc myFunc(&vbid, rg);
AmoebaMinimizer myMinimizer;
Vector startingPoint(2);
startingPoint[0] = 0; // pRefHet = 0.5
startingPoint[1] = -4.595; // pRefAlt = 0.01
myMinimizer.func = &myFunc;
myMinimizer.Reset(2);
myMinimizer.point = startingPoint;
myMinimizer.Minimize(1e-6);
double pRefHet = VerifyBamID::invLogit(myMinimizer.point[0]);
double pRefAlt = VerifyBamID::invLogit(myMinimizer.point[1]);
Logger::gLogger->writeLog("Reference Bias Estimated as ( Pr[refBase|HET] = %lf, Pr[refBase|ALT] = %lf) with LLK = %lf at readGroup %d",pRefHet,pRefAlt,myMinimizer.fmin,rg);
//vbid.setRefBiasParams(1.0, pRefHet, pRefAlt);
vbid.mixOut.llk0s[rg+1] = myFunc.llk0;
vbid.mixOut.llk1s[rg+1] = myFunc.llk1;
vbid.mixOut.refHets[rg+1] = myFunc.pRefHet;
vbid.mixOut.refAlts[rg+1] = myFunc.pRefAlt;
}
}
else if ( args.bFreeFull ) {
Logger::gLogger->writeLog("Performing chip-free estimation of reference-bias and sample mixture together");
for(int rg = -1; rg < vbid.nRGs - args.bIgnoreRG; ++rg) {
VerifyBamID::fullMixLLKFunc myFunc(&vbid, rg);
AmoebaMinimizer myMinimizer;
Vector startingPoint(3);
startingPoint[0] = -3.91; // start with fMix = 0.01
startingPoint[1] = 0; // pRefHet = 0.5
startingPoint[2] = -4.595; // pRefAlt = 0.01
myMinimizer.func = &myFunc;
myMinimizer.Reset(3);
myMinimizer.point = startingPoint;
myMinimizer.Minimize(1e-6);
double fMix = VerifyBamID::invLogit(myMinimizer.point[0]);
if ( fMix > 0.5 )
fMix = 1.-fMix;
double pRefHet = VerifyBamID::invLogit(myMinimizer.point[1]);
double pRefAlt = VerifyBamID::invLogit(myMinimizer.point[2]);
Logger::gLogger->writeLog("Optimal per-sample fMix = %lf\n",fMix);
Logger::gLogger->writeLog("Reference Bias Estimated as ( Pr[refBase|HET] = %lf, Pr[refBase|ALT] = %lf) with LLK = %lf",pRefHet,pRefAlt,myMinimizer.fmin);
//vbid.setRefBiasParams(1.0, pRefHet, pRefAlt);
vbid.mixOut.llk0s[rg+1] = myFunc.llk0;
vbid.mixOut.llk1s[rg+1] = myFunc.llk1;
vbid.mixOut.fMixs[rg+1] = myFunc.fMix;
vbid.mixOut.refHets[rg+1] = myFunc.pRefHet;
vbid.mixOut.refAlts[rg+1] = myFunc.pRefAlt;
}
}
Logger::gLogger->writeLog("calculating depth distribution");
vbid.calculateDepthDistribution(args.maxDepth, vbid.mixOut);
Logger::gLogger->writeLog("finished calculating depth distribution");
std::vector<int> bestInds(vbid.nRGs+1,-1);
std::vector<int> selfInds(vbid.nRGs+1,-1);
if ( args.bChipNone ) {
// do nothing
Logger::gLogger->writeLog("Skipping with-chip estimation of sample mixture");
}
else if ( args.bChipMixOnly ) {
Logger::gLogger->writeLog("Performing with-chip estimation of sample mixture at fixed reference bias parameter (%lf, %lf, %lf)",args.pRefRef,args.pRefHet,args.pRefAlt);
for(int rg=-1; rg < (vbid.nRGs - (int)args.bIgnoreRG); ++rg) {
double maxIBD = -1;
VerifyBamID::ibdLLK ibd(&vbid);
for(int i=0; i < (int)vbid.pGenotypes->indids.size(); ++i) {
double fIBD = ibd.OptimizeLLK(i, rg);
Logger::gLogger->writeLog("Comparing with individual %s.. Optimal fIBD = %lf, LLK0 = %lf, LLK1 = %lf for readgroup %d",vbid.pGenotypes->indids[i].c_str(),fIBD, ibd.llk0, ibd.llk1, rg);
if ( maxIBD < fIBD ) {
bestInds[rg+1] = i;
vbid.bestOut.llk0s[rg+1] = ibd.llk0;
vbid.bestOut.llk1s[rg+1] = ibd.llk1;
vbid.bestOut.fMixs[rg+1] = 1-ibd.fIBD;
maxIBD = ibd.fIBD;
}
if ( ( (rg < 0) && (vbid.pPile->sBamSMID == vbid.pGenotypes->indids[i] ) ) || ( ( rg >= 0 ) && ( vbid.pPile->vsSMIDs[rg] == vbid.pGenotypes->indids[i]) ) ) {
selfInds[rg+1] = i;
vbid.selfOut.llk0s[rg+1] = ibd.llk0;
vbid.selfOut.llk1s[rg+1] = ibd.llk1;
vbid.selfOut.fMixs[rg+1] = 1-ibd.fIBD;
}
}
if ( bestInds[rg+1] >= 0 ) {
Logger::gLogger->writeLog("Best Matching Individual is %s with IBD = %lf",vbid.pGenotypes->indids[bestInds[rg+1]].c_str(),maxIBD);
vbid.calculateDepthByGenotype(bestInds[rg+1],rg,vbid.bestOut);
}
if ( selfInds[rg+1] >= 0 ) {
Logger::gLogger->writeLog("Self Individual is %s with IBD = %lf",vbid.pGenotypes->indids[selfInds[rg+1]].c_str(),vbid.selfOut.fMixs[rg+1]);
vbid.calculateDepthByGenotype(selfInds[rg+1],rg,vbid.selfOut);
}
}
}
else if ( args.bChipRefBiasOnly ) {
Logger::gLogger->writeLog("Performing with-chip estimation of reference-bias without sample mixture");
if ( args.bSelfOnly ) {
for(int rg=-1; rg < (vbid.nRGs - (int)args.bIgnoreRG); ++rg) {
VerifyBamID::refBiasIbdLLKFunc myFunc(&vbid, rg);
AmoebaMinimizer myMinimizer;
Vector startingPoint(2);
startingPoint[0] = 0; // pRefHet = 0.5
startingPoint[1] = -4.595; // pRefAlt = 0.01
myMinimizer.func = &myFunc;
myMinimizer.Reset(2);
myMinimizer.point = startingPoint;
myMinimizer.Minimize(1e-6);
double pRefHet = VerifyBamID::invLogit(myMinimizer.point[0]);
double pRefAlt = VerifyBamID::invLogit(myMinimizer.point[1]);
Logger::gLogger->writeLog("Reference Bias Estimated as ( Pr[refBase|HET] = %lf, Pr[refBase|ALT] = %lf) with LLK = %lf",pRefHet,pRefAlt,myMinimizer.fmin);
//vbid.setRefBiasParams(1.0, pRefHet, pRefAlt);
vbid.selfOut.llk0s[rg+1] = myFunc.llk0;
vbid.selfOut.llk1s[rg+1] = myFunc.llk1;
vbid.selfOut.refHets[rg+1] = myFunc.pRefHet;
vbid.selfOut.refAlts[rg+1] = myFunc.pRefAlt;
vbid.calculateDepthByGenotype(0,rg,vbid.selfOut);
}
}
else {
Logger::gLogger->warning("--self must be set for --chip-refBias to work. Skipping..");
}
}
else if ( args.bChipFull ) {
Logger::gLogger->writeLog("Performing with-chip estimation of reference-bias and sample mixture together");
for(int rg=-1; rg < (vbid.nRGs - (int)args.bIgnoreRG); ++rg) {
double maxIBD = -1;
for(int i=0; i < (int)vbid.pGenotypes->indids.size(); ++i) {
VerifyBamID::fullIbdLLKFunc myFunc(&vbid,i,rg);
AmoebaMinimizer myMinimizer;
Vector startingPoint(3);
startingPoint[0] = 3.91; // start with fIBD = 0.99
startingPoint[1] = 0; // pRefHet = 0.5
startingPoint[2] = -4.595; // pRefAlt = 0.01
myMinimizer.func = &myFunc;
myFunc.indIdx = i;
myMinimizer.Reset(3);
myMinimizer.point = startingPoint;
myMinimizer.Minimize(1e-6);
double fIBD = VerifyBamID::invLogit(myMinimizer.point[0]);
double pRefHet = VerifyBamID::invLogit(myMinimizer.point[1]);
double pRefAlt = VerifyBamID::invLogit(myMinimizer.point[2]);
Logger::gLogger->writeLog("Comparing with individual %s.. Optimal fIBD = %lf, LLK0 = %lf, LLK1 = %lf for readgroup %d",vbid.pGenotypes->indids[i].c_str(), fIBD, myFunc.llk0, myFunc.llk1, rg);
//Logger::gLogger->writeLog("Optimal per-sample fIBD = %lf, ",fIBD);
Logger::gLogger->writeLog("Reference Bias Estimated as ( Pr[refBase|HET] = %lf, Pr[refBase|ALT] = %lf ) with LLK = %lf",pRefHet,pRefAlt,myMinimizer.fmin);
if ( maxIBD < fIBD ) {
bestInds[rg+1] = i;
maxIBD = fIBD;
vbid.bestOut.llk0s[rg+1] = myFunc.llk0;
vbid.bestOut.llk1s[rg+1] = myFunc.llk1;
vbid.bestOut.fMixs[rg+1] = 1.-myFunc.fIBD;
vbid.bestOut.refHets[rg+1] = myFunc.pRefHet;
vbid.bestOut.refAlts[rg+1] = myFunc.pRefAlt;
}
if ( ( (rg < 0) && (vbid.pPile->sBamSMID == vbid.pGenotypes->indids[i] ) ) || ( ( rg >= 0 ) && ( vbid.pPile->vsSMIDs[rg] == vbid.pGenotypes->indids[i]) ) ) {
selfInds[rg+1] = i;
vbid.selfOut.llk0s[rg+1] = myFunc.llk0;
vbid.selfOut.llk1s[rg+1] = myFunc.llk1;
vbid.selfOut.fMixs[rg+1] = 1.-myFunc.fIBD;
vbid.selfOut.refHets[rg+1] = myFunc.pRefHet;
vbid.selfOut.refAlts[rg+1] = myFunc.pRefAlt;
vbid.calculateDepthByGenotype(i, rg, vbid.selfOut);
}
}
//vbid.setRefBiasParams(1.0, pRefHet, pRefAlt);
if ( bestInds[rg+1] >= 0 ) {
Logger::gLogger->writeLog("Best Matching Individual is %s with IBD = %lf",vbid.pGenotypes->indids[bestInds[rg+1]].c_str(),maxIBD);
vbid.calculateDepthByGenotype(bestInds[rg+1], rg, vbid.bestOut);
}
if ( selfInds[rg+1] >= 0 ) {
Logger::gLogger->writeLog("Self Individual is %s with IBD = %lf",vbid.pGenotypes->indids[selfInds[rg+1]].c_str(),vbid.selfOut.fMixs[rg+1]);
vbid.calculateDepthByGenotype(selfInds[rg+1],rg,vbid.selfOut);
}
}
}
// PRINT OUTPUT FILE - ".selfSM"
// [SEQ_ID] : SAMPLE ID in the sequence file
// [CHIP_ID] : SAMPLE ID in the chip file (NA if not available)
// [#SNPS] : Number of markers evaluated
// [#READS] : Number of reads evaluated
// [AVG_DP] : Mean depth
// [FREEMIX] : Chip-free estimated alpha (% MIX in 0-1 scale), NA if unavailable
// [FREELK1] : Chip-free log-likelihood at estimated alpha
// [FREELK0] : Chip-free log-likelihood at 0% contamination
// [CHIPIBD] : With-chip estimated alpha (% MIX in 0-1 scale)
// [CHIPLK1] : With-chip log-likelihood at estimated alpha
// [CHIPLK0] : With-chip log-likelihood at 0% contamination
// [DPREF] : Depth at reference site in the chip
// [RDPHET] : Relative depth at HET site in the chip
// [RDPALT] : Relative depth at HOMALT site in the chip
// [FREE_RF] : Pr(Ref|Ref) site estimated without chip data
// [FREE_RH] : Pr(Ref|Het) site estimated without chip data
// [FREE_RA] : Pr(Ref|Alt) site estimated without chip data
// [CHIP_RF] : Pr(Ref|Ref) site estimated with chip data
// [CHIP_RH] : Pr(Ref|Het) site estimated with chip data
// [CHIP_RA] : Pr(Ref|Alt) site estimated with chip data
// [DPREF] : Depth at reference alleles
// [RDPHET] : Relative depth at heterozygous alleles
// [RDPALT] : Relative depth at hom-alt alleles
String selfSMFN = args.sOutFile + ".selfSM";
String bestSMFN = args.sOutFile + ".bestSM";
String selfRGFN = args.sOutFile + ".selfRG";
String bestRGFN = args.sOutFile + ".bestRG";
String dpSMFN = args.sOutFile + ".depthSM";
String dpRGFN = args.sOutFile + ".depthRG";
IFILE selfSMF = ifopen(selfSMFN,"wb");
IFILE bestSMF = (args.bFindBest ? ifopen(bestSMFN,"wb") : NULL);
IFILE selfRGF = (args.bIgnoreRG ? NULL : ifopen(selfRGFN,"wb"));
IFILE bestRGF = (args.bFindBest && !args.bIgnoreRG) ? ifopen(bestRGFN,"wb") : NULL;
IFILE dpSMF = ifopen(dpSMFN,"wb");
IFILE dpRGF = (args.bIgnoreRG ? NULL : ifopen(dpRGFN,"wb"));
if ( selfSMF == NULL ) {
Logger::gLogger->error("Cannot write to %s",selfSMF);
}
if ( args.bFindBest && ( bestSMF == NULL ) ) {
Logger::gLogger->error("Cannot write to %s",bestSMF);
}
if ( dpSMF == NULL ) {
Logger::gLogger->error("Cannot write to %s",dpSMF);
}
ifprintf(dpSMF,"#RG\tDEPTH\t#SNPs\t%%SNPs\t%%CUMUL\n");
int nCumMarkers = 0;
for(int i=args.maxDepth; i >= 0; --i) {
nCumMarkers += vbid.mixOut.depths[i];
ifprintf(dpSMF,"ALL\t%d\t%d\t%.5lf\t%.5lf\n",i, vbid.mixOut.depths[i],(double) vbid.mixOut.depths[i]/(double)vbid.nMarkers,(double)nCumMarkers/(double)vbid.nMarkers);
}
ifclose(dpSMF);
if ( dpRGF != NULL ) {
ifprintf(dpRGF,"#RG\tDEPTH\t#SNPs\t%%SNPs\t%%CUMUL\n");
for(int rg=0; rg < (vbid.nRGs - (int)args.bIgnoreRG); ++rg) {
const char* rgID = vbid.pPile->vsRGIDs[rg].c_str();
int nMarkers = 0;
for(int i=args.maxDepth; i >= 0; --i) {
nMarkers += vbid.mixOut.depths[(rg+1)*(args.maxDepth+1) + i];
}
nCumMarkers = 0;
for(int i=args.maxDepth; i >= 0; --i) {
int d = vbid.mixOut.depths[(rg+1)*(args.maxDepth+1) + i];
nCumMarkers += d;
ifprintf(dpRGF,"%s\t%d\t%d\t%.5lf\t%.5lf\n",rgID,i,d,(double)d/(double)vbid.nMarkers,(double)nCumMarkers/(double)nMarkers);
}
}
ifclose(dpRGF);
}
const char* headers[] = {"#SEQ_ID","RG","CHIP_ID","#SNPS","#READS","AVG_DP","FREEMIX","FREELK1","FREELK0","FREE_RH","FREE_RA","CHIPMIX","CHIPLK1","CHIPLK0","CHIP_RH","CHIP_RA","DPREF","RDPHET","RDPALT"};
int nheaders = sizeof(headers)/sizeof(headers[0]);
for(int i=0; i < nheaders; ++i) { ifprintf(selfSMF,"%s%s",i>0 ? "\t" : "",headers[i]); }
ifprintf(selfSMF,"\n");
ifprintf(selfSMF,"%s\tALL",vbid.pPile->sBamSMID.c_str());
ifprintf(selfSMF,"\t%s",selfInds[0] >= 0 ? vbid.pGenotypes->indids[selfInds[0]].c_str() : "NA");
ifprintf(selfSMF,"\t%d\t%d\t%.2lf",vbid.nMarkers,vbid.mixOut.numReads[0],(double)vbid.mixOut.numReads[0]/(double)vbid.nMarkers);
if ( args.bFreeNone ) { ifprintf(selfSMF,"\tNA\tNA\tNA\tNA\tNA"); }
else if ( args.bFreeMixOnly ) { ifprintf(selfSMF,"\t%.5lf\t%.2lf\t%.2lf\tNA\tNA",vbid.mixOut.fMixs[0],vbid.mixOut.llk1s[0],vbid.mixOut.llk0s[0]); }
else if ( args.bFreeRefBiasOnly ) { ifprintf(selfSMF,"\tNA\t%.2lf\t%.2lf\t%.5lf\t%.5lf",vbid.mixOut.llk1s[0],vbid.mixOut.llk0s[0],vbid.mixOut.refHets[0],vbid.mixOut.refAlts[0]); }
else if ( args.bFreeFull ) { ifprintf(selfSMF,"\t%.5lf\t%.2lf\t%.2lf\t%.5lf\t%.5lf",vbid.mixOut.fMixs[0],vbid.mixOut.llk1s[0],vbid.mixOut.llk0s[0],vbid.mixOut.refHets[0],vbid.mixOut.refAlts[0]); }
else { error("Invalid option in handling bFree"); }
if ( args.bChipNone || bestInds[0] < 0 ) { ifprintf(selfSMF,"\tNA\tNA\tNA\tNA\tNA\tNA\tNA\tNA"); }
else if ( args.bChipMixOnly ) { ifprintf(selfSMF,"\t%.5lf\t%.2lf\t%.2lf\tNA\tNA\t%.3lf\t%.4lf\t%.4lf",vbid.selfOut.fMixs[0],vbid.selfOut.llk1s[0],vbid.selfOut.llk0s[0],(double)vbid.selfOut.numReads[1]/vbid.selfOut.numGenos[1], (double)vbid.selfOut.numReads[2]*vbid.selfOut.numGenos[1]/vbid.selfOut.numReads[1]/vbid.selfOut.numGenos[2], (double)vbid.selfOut.numReads[3]*vbid.selfOut.numGenos[1]/vbid.selfOut.numReads[1]/vbid.selfOut.numGenos[3]); }
else if ( args.bChipMixOnly ) { ifprintf(selfSMF,"\tNA\t%.2lf\t%.2lf\t%.5lf\t%.5lf\t%.3lf\t%.4lf\t%.4lf",vbid.selfOut.llk1s[0], vbid.selfOut.llk0s[0], vbid.selfOut.refHets[0], vbid.selfOut.refAlts[0], (double)vbid.selfOut.numReads[1]/vbid.selfOut.numGenos[1], (double)vbid.selfOut.numReads[2]*vbid.selfOut.numGenos[1]/vbid.selfOut.numReads[1]/vbid.selfOut.numGenos[2], (double)vbid.selfOut.numReads[3]*vbid.selfOut.numGenos[1]/vbid.selfOut.numReads[1]/vbid.selfOut.numGenos[3]); }
else if ( args.bChipFull ) { ifprintf(selfSMF,"\t%.5lf\t%.2lf\t%.2lf\t%.5lf\t%.5lf\t%.3lf\t%.4lf\t%.4lf", vbid.selfOut.fMixs[0], vbid.selfOut.llk1s[0], vbid.selfOut.llk0s[0], vbid.selfOut.refHets[0], vbid.selfOut.refAlts[0], (double)vbid.selfOut.numReads[1]/vbid.selfOut.numGenos[1], (double)vbid.selfOut.numReads[2]*vbid.selfOut.numGenos[1]/vbid.selfOut.numReads[1]/vbid.selfOut.numGenos[2], (double)vbid.selfOut.numReads[3]*vbid.selfOut.numGenos[1]/vbid.selfOut.numReads[1]/vbid.selfOut.numGenos[3]); }
else { error("Invalid option in handling bChip"); }
ifprintf(selfSMF,"\n");
ifclose(selfSMF);
if ( bestSMF != NULL ) {
for(int i=0; i < nheaders; ++i) { ifprintf(bestSMF,"%s%s",i>0 ? "\t" : "",headers[i]); }
ifprintf(bestSMF,"\n");
ifprintf(bestSMF,"%s\tALL",vbid.pPile->sBamSMID.c_str());
ifprintf(bestSMF,"\t%s",bestInds[0] >= 0 ? vbid.pGenotypes->indids[bestInds[0]].c_str() : "NA");
ifprintf(bestSMF,"\t%d\t%d\t%.2lf",vbid.nMarkers,vbid.mixOut.numReads[0],(double)vbid.mixOut.numReads[0]/(double)vbid.nMarkers);
if ( args.bFreeNone ) { ifprintf(bestSMF,"\tNA\tNA\tNA\tNA\tNA"); }
else if ( args.bFreeMixOnly ) { ifprintf(bestSMF,"\t%.5lf\t%.2lf\t%.2lf\tNA\tNA",vbid.mixOut.fMixs[0],vbid.mixOut.llk1s[0],vbid.mixOut.llk0s[0]); }
else if ( args.bFreeRefBiasOnly ) { ifprintf(bestSMF,"\tNA\t%.2lf\t%.2lf\t%.5lf\t%.5lf",vbid.mixOut.llk1s[0],vbid.mixOut.llk0s[0],vbid.mixOut.refHets[0],vbid.mixOut.refAlts[0]); }
else if ( args.bFreeFull ) { ifprintf(bestSMF,"\t%.5lf\t%.2lf\t%.2lf\t%.5lf\t%.5lf",vbid.mixOut.fMixs[0],vbid.mixOut.llk1s[0],vbid.mixOut.llk0s[0],vbid.mixOut.refHets[0],vbid.mixOut.refAlts[0]); }
else { error("Invalid option in handling bFree"); }
if ( args.bChipNone || bestInds[0] < 0 ) { ifprintf(bestSMF,"\tNA\tNA\tNA\tNA\tNA\tNA\tNA\tNA"); }
else if ( args.bChipMixOnly ) { ifprintf(bestSMF,"\t%.5lf\t%.2lf\t%.2lf\tNA\tNA\t%.3lf\t%.4lf\t%.4lf",vbid.bestOut.fMixs[0],vbid.bestOut.llk1s[0],vbid.bestOut.llk0s[0],(double)vbid.bestOut.numReads[1]/vbid.bestOut.numGenos[1], (double)vbid.bestOut.numReads[2]*vbid.bestOut.numGenos[1]/vbid.bestOut.numReads[1]/vbid.bestOut.numGenos[2], (double)vbid.bestOut.numReads[3]*vbid.bestOut.numGenos[1]/vbid.bestOut.numReads[1]/vbid.bestOut.numGenos[3]); }
else if ( args.bChipMixOnly ) { ifprintf(bestSMF,"\tNA\t%.2lf\t%.2lf\t%.5lf\t%.5lf\t%.3lf\t%.4lf\t%.4lf",vbid.bestOut.llk1s[0], vbid.bestOut.llk0s[0], vbid.bestOut.refHets[0], vbid.bestOut.refAlts[0], (double)vbid.bestOut.numReads[1]/vbid.bestOut.numGenos[1], (double)vbid.bestOut.numReads[2]*vbid.bestOut.numGenos[1]/vbid.bestOut.numReads[1]/vbid.bestOut.numGenos[2], (double)vbid.bestOut.numReads[3]*vbid.bestOut.numGenos[1]/vbid.bestOut.numReads[1]/vbid.bestOut.numGenos[3]); }
else if ( args.bChipFull ) { ifprintf(bestSMF,"\t%.5lf\t%.2lf\t%.2lf\t%.5lf\t%.5lf\t%.3lf\t%.4lf\t%.4lf", vbid.bestOut.fMixs[0], vbid.bestOut.llk1s[0], vbid.bestOut.llk0s[0], vbid.bestOut.refHets[0], vbid.bestOut.refAlts[0], (double)vbid.bestOut.numReads[1]/vbid.bestOut.numGenos[1], (double)vbid.bestOut.numReads[2]*vbid.bestOut.numGenos[1]/vbid.bestOut.numReads[1]/vbid.bestOut.numGenos[2], (double)vbid.bestOut.numReads[3]*vbid.bestOut.numGenos[1]/vbid.bestOut.numReads[1]/vbid.bestOut.numGenos[3]); }
else { error("Invalid option in handling bChip"); }
ifprintf(bestSMF,"\n");
ifclose(bestSMF);
}
if ( selfRGF != NULL ) {
for(int i=0; i < nheaders; ++i) { ifprintf(selfRGF,"%s%s",i>0 ? "\t" : "",headers[i]); }
ifprintf(selfRGF,"\n");
for(int rg=0; rg < vbid.nRGs; ++rg) {
ifprintf(selfRGF,"%s\t%s",vbid.pPile->sBamSMID.c_str(),vbid.pPile->vsRGIDs[rg].c_str());
ifprintf(selfRGF,"\t%s",bestInds[rg] >= 0 ? vbid.pGenotypes->indids[bestInds[rg]].c_str() : "NA");
ifprintf(selfRGF,"\t%d\t%d\t%.2lf",vbid.nMarkers,vbid.mixOut.numReads[(rg+1)*4],(double)vbid.mixOut.numReads[(rg+1)*4]/(double)vbid.mixOut.numGenos[(rg+1)*4]);
if ( args.bFreeNone ) { ifprintf(selfRGF,"\tNA\tNA\tNA\tNA\tNA"); }
else if ( args.bFreeMixOnly ) { ifprintf(selfRGF,"\t%.5lf\t%.2lf\t%.2lf\tNA\tNA",vbid.mixOut.fMixs[rg+1],vbid.mixOut.llk1s[rg+1],vbid.mixOut.llk0s[rg+1]); }
else if ( args.bFreeRefBiasOnly ) { ifprintf(selfRGF,"\tNA\t%.2lf\t%.2lf\t%.5lf\t%.5lf",vbid.mixOut.llk1s[rg+1],vbid.mixOut.llk0s[rg+1],vbid.mixOut.refHets[rg+1],vbid.mixOut.refAlts[rg+1]); }
else if ( args.bFreeFull ) { ifprintf(selfRGF,"\t%.5lf\t%.2lf\t%.2lf\t%.5lf\t%.5lf",vbid.mixOut.fMixs[rg+1],vbid.mixOut.llk1s[rg+1],vbid.mixOut.llk0s[rg+1],vbid.mixOut.refHets[rg+1],vbid.mixOut.refAlts[rg+1]); }
else { error("Invalid option in handling bFree"); }
if ( args.bChipNone || bestInds[0] < 0 ) { ifprintf(selfRGF,"\tNA\tNA\tNA\tNA\tNA\tNA\tNA\tNA"); }
else if ( args.bChipMixOnly ) { ifprintf(selfRGF,"\t%.5lf\t%.2lf\t%.2lf\tNA\tNA\t%.3lf\t%.4lf\t%.4lf",vbid.selfOut.fMixs[rg+1], vbid.selfOut.llk1s[rg+1], vbid.selfOut.llk0s[rg+1], (double)vbid.selfOut.numReads[(rg+1)*4+1]/vbid.selfOut.numGenos[(rg+1)*4+1], (double)vbid.selfOut.numReads[(rg+1)*4+2]*vbid.selfOut.numGenos[(rg+1)*4+1]/vbid.selfOut.numReads[(rg+1)*4+1]/vbid.selfOut.numGenos[(rg+1)*4+2], (double)vbid.selfOut.numReads[(rg+1)*4+3]*vbid.selfOut.numGenos[(rg+1)*4+1]/vbid.selfOut.numReads[(rg+1)*4+1]/vbid.selfOut.numGenos[(rg+1)*4+3]); }
else if ( args.bChipMixOnly ) { ifprintf(selfRGF,"\tNA\t%.2lf\t%.2lf\t%.5lf\t%.5lf\t%.3lf\t%.4lf\t%.4lf",vbid.selfOut.llk1s[rg+1], vbid.selfOut.llk0s[rg+1], vbid.selfOut.refHets[rg+1], vbid.selfOut.refAlts[rg+1], (double)vbid.selfOut.numReads[(rg+1)*4+1]/vbid.selfOut.numGenos[(rg+1)*4+1], (double)vbid.selfOut.numReads[(rg+1)*4+2]*vbid.selfOut.numGenos[(rg+1)*4+1]/vbid.selfOut.numReads[(rg+1)*4]/vbid.selfOut.numGenos[(rg+1)*4+2], (double)vbid.selfOut.numReads[(rg+1)*4+3]*vbid.selfOut.numGenos[(rg+1)*4+1]/vbid.selfOut.numReads[(rg+1)*4+1]/vbid.selfOut.numGenos[(rg+1)*4+3]); }
else if ( args.bChipFull ) { ifprintf(selfRGF,"\t%.5lf\t%.2lf\t%.2lf\t%.5lf\t%.5lf\t%.3lf\t%.4lf\t%.4lf", vbid.selfOut.fMixs[rg+1], vbid.selfOut.llk1s[rg+1], vbid.selfOut.llk0s[rg+1], vbid.selfOut.refHets[rg+1], vbid.selfOut.refAlts[rg+1], (double)vbid.selfOut.numReads[(rg+1)*4+1]/vbid.selfOut.numGenos[(rg+1)*4+1], (double)vbid.selfOut.numReads[(rg+1)*4+2]*vbid.selfOut.numGenos[(rg+1)*4+1]/vbid.selfOut.numReads[(rg+1)*4+1]/vbid.selfOut.numGenos[(rg+1)*4+2], (double)vbid.selfOut.numReads[(rg+1)*4+3]*vbid.selfOut.numGenos[(rg+1)*4+1]/vbid.selfOut.numReads[(rg+1)*4+1]/vbid.selfOut.numGenos[(rg+1)*4+3]); }
else { error("Invalid option in handling bChip"); }
ifprintf(selfRGF,"\n");
}
ifclose(selfRGF);
}
if ( bestRGF != NULL ) {
for(int i=0; i < nheaders; ++i) { ifprintf(bestRGF,"%s%s",i>0 ? "\t" : "",headers[i]); }
ifprintf(bestRGF,"\n");
for(int rg=0; rg < vbid.nRGs; ++rg) {
ifprintf(bestRGF,"%s\t%s",vbid.pPile->sBamSMID.c_str(),vbid.pPile->vsRGIDs[rg].c_str());
ifprintf(bestRGF,"\t%s",bestInds[rg] >= 0 ? vbid.pGenotypes->indids[bestInds[rg]].c_str() : "NA");
ifprintf(bestRGF,"\t%d\t%d\t%.2lf",vbid.nMarkers,vbid.mixOut.numReads[(rg+1)*4],(double)vbid.mixOut.numReads[(rg+1)*4]/(double)vbid.mixOut.numGenos[(rg+1)*4]);
if ( args.bFreeNone ) { ifprintf(bestRGF,"\tNA\tNA\tNA\tNA\tNA"); }
else if ( args.bFreeMixOnly ) { ifprintf(bestRGF,"\t%.5lf\t%.2lf\t%.2lf\tNA\tNA",vbid.mixOut.fMixs[rg+1],vbid.mixOut.llk1s[rg+1],vbid.mixOut.llk0s[rg+1]); }
else if ( args.bFreeRefBiasOnly ) { ifprintf(bestRGF,"\tNA\t%.2lf\t%.2lf\t%.5lf\t%.5lf",vbid.mixOut.llk1s[rg+1],vbid.mixOut.llk0s[rg+1],vbid.mixOut.refHets[rg+1],vbid.mixOut.refAlts[rg+1]); }
else if ( args.bFreeFull ) { ifprintf(bestRGF,"\t%.5lf\t%.2lf\t%.2lf\t%.5lf\t%.5lf",vbid.mixOut.fMixs[rg+1],vbid.mixOut.llk1s[rg+1],vbid.mixOut.llk0s[rg+1],vbid.mixOut.refHets[rg+1],vbid.mixOut.refAlts[rg+1]); }
else { error("Invalid option in handling bFree"); }
if ( args.bChipNone || bestInds[0] < 0 ) { ifprintf(bestRGF,"\tNA\tNA\tNA\tNA\tNA\tNA\tNA\tNA"); }
else if ( args.bChipMixOnly ) { ifprintf(bestRGF,"\t%.5lf\t%.2lf\t%.2lf\tNA\tNA\t%.3lf\t%.4lf\t%.4lf",vbid.bestOut.fMixs[rg+1], vbid.bestOut.llk1s[rg+1], vbid.bestOut.llk0s[rg+1], (double)vbid.bestOut.numReads[(rg+1)*4+1]/vbid.bestOut.numGenos[(rg+1)*4+1], (double)vbid.bestOut.numReads[(rg+1)*4+2]*vbid.bestOut.numGenos[(rg+1)*4+1]/vbid.bestOut.numReads[(rg+1)*4+1]/vbid.bestOut.numGenos[(rg+1)*4+2], (double)vbid.bestOut.numReads[(rg+1)*4+3]*vbid.bestOut.numGenos[(rg+1)*4+1]/vbid.bestOut.numReads[(rg+1)*4+1]/vbid.bestOut.numGenos[(rg+1)*4+3]); }
else if ( args.bChipMixOnly ) { ifprintf(bestRGF,"\tNA\t%.2lf\t%.2lf\t%.5lf\t%.5lf\t%.3lf\t%.4lf\t%.4lf",vbid.bestOut.llk1s[rg+1], vbid.bestOut.llk0s[rg+1], vbid.bestOut.refHets[rg+1], vbid.bestOut.refAlts[rg+1], (double)vbid.bestOut.numReads[(rg+1)*4+1]/vbid.bestOut.numGenos[(rg+1)*4+1], (double)vbid.bestOut.numReads[(rg+1)*4+2]*vbid.bestOut.numGenos[(rg+1)*4+1]/vbid.bestOut.numReads[(rg+1)*4]/vbid.bestOut.numGenos[(rg+1)*4+2], (double)vbid.bestOut.numReads[(rg+1)*4+3]*vbid.bestOut.numGenos[(rg+1)*4+1]/vbid.bestOut.numReads[(rg+1)*4+1]/vbid.bestOut.numGenos[(rg+1)*4+3]); }
else if ( args.bChipFull ) { ifprintf(bestRGF,"\t%.5lf\t%.2lf\t%.2lf\t%.5lf\t%.5lf\t%.3lf\t%.4lf\t%.4lf", vbid.bestOut.fMixs[rg+1], vbid.bestOut.llk1s[rg+1], vbid.bestOut.llk0s[rg+1], vbid.bestOut.refHets[rg+1], vbid.bestOut.refAlts[rg+1], (double)vbid.bestOut.numReads[(rg+1)*4+1]/vbid.bestOut.numGenos[(rg+1)*4+1], (double)vbid.bestOut.numReads[(rg+1)*4+2]*vbid.bestOut.numGenos[(rg+1)*4+1]/vbid.bestOut.numReads[(rg+1)*4+1]/vbid.bestOut.numGenos[(rg+1)*4+2], (double)vbid.bestOut.numReads[(rg+1)*4+3]*vbid.bestOut.numGenos[(rg+1)*4+1]/vbid.bestOut.numReads[(rg+1)*4+1]/vbid.bestOut.numGenos[(rg+1)*4+3]); }
else { error("Invalid option in handling bChip"); }
ifprintf(bestRGF,"\n");
}
ifclose(bestRGF);
}
time(&t);
Logger::gLogger->writeLog("Analysis finished on %s",ctime(&t));
return 0;
}
int VcfMac::execute(int argc, char **argv)
{
String inputVcf = "";
int minAC = -1;
String sampleSubset = "";
String filterList = "";
bool params = false;
IntervalTree<int> regions;
std::vector<int> intersection;
// Read in the parameters.
ParameterList inputParameters;
BEGIN_LONG_PARAMETERS(longParameterList)
LONG_PARAMETER_GROUP("Required Parameters")
LONG_STRINGPARAMETER("in", &inputVcf)
LONG_PARAMETER_GROUP("Optional Parameters")
LONG_STRINGPARAMETER("sampleSubset", &sampleSubset)
LONG_INTPARAMETER("minAC", &minAC)
LONG_STRINGPARAMETER("filterList", &filterList)
LONG_PARAMETER("params", ¶ms)
LONG_PHONEHOME(VERSION)
END_LONG_PARAMETERS();
inputParameters.Add(new LongParameters ("Input Parameters",
longParameterList));
inputParameters.Read(argc-1, &(argv[1]));
// Check that all files were specified.
if(inputVcf == "")
{
usage();
inputParameters.Status();
std::cerr << "Missing \"--in\", a required parameter.\n\n";
return(-1);
}
if(params)
{
inputParameters.Status();
}
// Open the two input files.
VcfFileReader inFile;
VcfHeader header;
VcfRecord record;
// Open the file
if(sampleSubset.IsEmpty())
{
inFile.open(inputVcf, header);
}
else
{
inFile.open(inputVcf, header, sampleSubset, NULL, NULL);
}
// Add the discard rule for minor allele count.
if(minAC >= 0)
{
inFile.addDiscardMinMinorAlleleCount(minAC, NULL);
}
if(!filterList.IsEmpty())
{
// Open the filter list.
IFILE regionFile = ifopen(filterList, "r");
String regionLine;
StringArray regionColumn;
int start;
int end;
int intervalVal = 1;
if(regionFile == NULL)
{
std::cerr << "Failed to open " << filterList
<< ", so keeping all positions\n";
filterList.Clear();
}
else
{
while( regionFile->isOpen() && !regionFile->ifeof())
{
// Read the next interval
regionLine.Clear();
regionLine.ReadLine(regionFile);
if(regionLine.IsEmpty())
{
// Nothing on this line, continue to the next.
continue;
}
regionColumn.ReplaceColumns(regionLine, ' ');
if(regionColumn.Length() != 2)
{
std::cerr << "Improperly formatted region line: "
<< regionLine << "; skipping to the next line.\n";
continue;
}
// Convert the columns to integers.
if(!regionColumn[0].AsInteger(start))
{
// The start position (1st column) is not an integer.
std::cerr << "Improperly formatted region line, start position "
<< "(1st column) is not an integer: "
<< regionColumn[0]
<< "; Skipping to the next line.\n";
continue;
}
if(!regionColumn[1].AsInteger(end))
{
// The start position (1st column) is not an integer.
std::cerr << "Improperly formatted region line, end position "
<< "(2nd column) is not an integer: "
<< regionColumn[1]
<< "; Skipping to the next line.\n";
continue;
}
// Add 1-based inclusive intervals.
regions.add(start,end, intervalVal);
}
}
}
int numReadRecords = 0;
while( inFile.readRecord(record))
{
if(!filterList.IsEmpty())
{
// Check if the region should be kept.
intersection.clear();
regions.get_intersecting_intervals(record.get1BasedPosition(), intersection);
if(intersection.empty())
{
// not in the interval, so continue to the next record.
continue;
}
}
++numReadRecords;
// Loop through the number of possible alternates.
unsigned int numAlts = record.getNumAlts();
int minAlleleCount = -1;
int curAlleleCount = 0;
int totalAlleleCount = 0;
for(unsigned int i = 0; i <= numAlts; i++)
{
curAlleleCount = record.getAlleleCount(i);
if((minAlleleCount == -1) ||
(curAlleleCount < minAlleleCount))
{
minAlleleCount = curAlleleCount;
}
totalAlleleCount += curAlleleCount;
}
if(totalAlleleCount != 0)
{
double maf = (double)minAlleleCount/totalAlleleCount;
std::cout << record.getIDStr()
<< "\t" << minAlleleCount
<< "\t" << maf << "\n";
}
}
inFile.close();
// std::cerr << "\n\t# Records: " << numReadRecords << "\n";
// return success.
return(0);
}
// Dump the reference information from specified SAM/BAM file.
int DumpRefInfo::execute(int argc, char **argv)
{
// Extract command line arguments.
String inFile = "";
bool noeof = false;
bool printRecordRefs = false;
bool params = false;
ParameterList inputParameters;
BEGIN_LONG_PARAMETERS(longParameterList)
LONG_STRINGPARAMETER("in", &inFile)
LONG_PARAMETER("noeof", &noeof)
LONG_PARAMETER("printRecordRefs", &printRecordRefs)
LONG_PARAMETER("params", ¶ms)
LONG_PHONEHOME(VERSION)
END_LONG_PARAMETERS();
inputParameters.Add(new LongParameters ("Input Parameters",
longParameterList));
// parameters start at index 2 rather than 1.
inputParameters.Read(argc, argv, 2);
// If no eof block is required for a bgzf file, set the bgzf file type to
// not look for it.
if(noeof)
{
// Set that the eof block is not required.
BgzfFileType::setRequireEofBlock(false);
}
// Check to see if the in file was specified, if not, report an error.
if(inFile == "")
{
usage();
inputParameters.Status();
// In file was not specified but it is mandatory.
std::cerr << "--in is a mandatory argument, "
<< "but was not specified" << std::endl;
return(-1);
}
if(params)
{
inputParameters.Status();
}
// Open the input file for reading.
SamFile samIn;
samIn.OpenForRead(inFile);
// Read the sam header.
SamFileHeader samHeader;
samIn.ReadHeader(samHeader);
const SamReferenceInfo& refInfo = samHeader.getReferenceInfo();
int numReferences = refInfo.getNumEntries();
for(int i = 0; i < numReferences; i++)
{
std::cout << "Reference Index " << i;
std::cout << "; Name: " << refInfo.getReferenceName(i)
<< std::endl;
}
if(numReferences == 0)
{
// There is no reference info.
std::cerr << "The header contains no reference information.\n";
}
// If we are to print the references as found in the records, loop
// through reading the records.
if(printRecordRefs)
{
SamRecord samRecord;
// Track the prev name/id.
std::string prevName = "";
int prevID = -2;
int recCount = 0; // track the num records in a ref.
// Keep reading records until ReadRecord returns false.
while(samIn.ReadRecord(samHeader, samRecord))
{
const char* name = samRecord.getReferenceName();
int id = samRecord.getReferenceID();
if((strcmp(name, prevName.c_str()) != 0) || (id != prevID))
{
if(prevID != -2)
{
std::cout << "\tRef ID: " << prevID
<< "\tRef Name: " << prevName
<< "\tNumRecs: " << recCount
<< std::endl;
}
recCount = 0;
prevID = id;
prevName = name;
}
++recCount;
}
// Print the last index.
if(prevID != -2)
{
std::cout << "\tRef ID: " << prevID
<< "\tRef Name: " << prevName
<< "\tNumRecs: " << recCount
<< std::endl;
}
}
return(SamStatus::SUCCESS);
}
