SamFile::SortedType SamFile::getSortOrderFromHeader(SamFileHeader& header) { const char* tag = header.getSortOrder(); // Default to unsorted since if it is not specified in the header // that is the value that should be used. SortedType headerSortOrder = UNSORTED; if(strcmp(tag, "queryname") == 0) { headerSortOrder = QUERY_NAME; } else if(strcmp(tag, "coordinate") == 0) { headerSortOrder = COORDINATE; } return(headerSortOrder); }
int Bam2FastQ::execute(int argc, char **argv) { // Extract command line arguments. String inFile = ""; bool readName = false; String refFile = ""; String outBase = ""; String firstOut = ""; String secondOut = ""; String unpairedOut = ""; bool interleave = false; bool noeof = false; bool params = false; myNumMateFailures = 0; myNumPairs = 0; myNumUnpaired = 0; myReverseComp = true; myRNPlus = false; myFirstRNExt = DEFAULT_FIRST_EXT; mySecondRNExt = DEFAULT_SECOND_EXT; 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("merge", &interleave) LONG_STRINGPARAMETER("refFile", &refFile) LONG_STRINGPARAMETER("firstRNExt", &myFirstRNExt) LONG_STRINGPARAMETER("secondRNExt", &mySecondRNExt) LONG_PARAMETER("rnPlus", &myRNPlus) LONG_PARAMETER("noReverseComp", &myReverseComp) LONG_PARAMETER("noeof", &noeof) LONG_PARAMETER("params", ¶ms) LONG_PARAMETER_GROUP("Optional OutputFile Names") LONG_STRINGPARAMETER("outBase", &outBase) 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); } // 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); } // Check to see if the out file was specified, if not, generate it from // the input filename. if(outBase == "") { // Just remove the extension from the input filename. int extStart = inFile.FastFindLastChar('.'); if(extStart <= 0) { outBase = inFile; } else { outBase = inFile.Left(extStart); } } // Check to see if the first/second/single-ended were specified and // if not, set them. std::string firstExt = "_1.fastq"; if(interleave) { firstExt = "_interleaved.fastq"; } getFileName(firstOut, outBase, firstExt.c_str()); getFileName(secondOut, outBase, "_2.fastq"); getFileName(unpairedOut, outBase, ".fastq"); 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; SamFileHeader samHeader; samIn.OpenForRead(inFile, &samHeader); // Open the output files. myUnpairedFile = ifopen(unpairedOut, "w"); // Only open the first file if it is different than an already opened file. if(firstOut != unpairedOut) { myFirstFile = ifopen(firstOut, "w"); } 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"); } 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(samHeader.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(samHeader, *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); } } // 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"; } return(returnStatus); }