X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=chimeraslayercommand.cpp;h=1dd3d4221975a33bf52e5e6031029c97d66e0ca5;hb=9013e13ecfb2fda3c2664a76f76cc99b8c7fa74c;hp=336dba939f1fd114c6ace70f1ddc2501448df336;hpb=fdc1f6eaf544f695fc1511f24bddd7e6069c33ba;p=mothur.git diff --git a/chimeraslayercommand.cpp b/chimeraslayercommand.cpp index 336dba9..1dd3d42 100644 --- a/chimeraslayercommand.cpp +++ b/chimeraslayercommand.cpp @@ -8,10 +8,6 @@ */ #include "chimeraslayercommand.h" -#include "bellerophon.h" -#include "pintail.h" -#include "ccode.h" -#include "chimeracheckrdp.h" #include "chimeraslayer.h" @@ -33,7 +29,7 @@ ChimeraSlayerCommand::ChimeraSlayerCommand(string option) { OptionParser parser(option); map parameters = parser.getParameters(); - ValidParameters validParameter; + ValidParameters validParameter("chimera.slayer"); map::iterator it; //check to make sure all parameters are valid for command @@ -46,14 +42,6 @@ ChimeraSlayerCommand::ChimeraSlayerCommand(string option) { if (inputDir == "not found"){ inputDir = ""; } else { string path; - it = parameters.find("fasta"); - //user has given a template file - if(it != parameters.end()){ - path = hasPath(it->second); - //if the user has not given a path then, add inputdir. else leave path alone. - if (path == "") { parameters["fasta"] = inputDir + it->second; } - } - it = parameters.find("template"); //user has given a template file if(it != parameters.end()){ @@ -65,9 +53,65 @@ ChimeraSlayerCommand::ChimeraSlayerCommand(string option) { //check for required parameters - fastafile = validParameter.validFile(parameters, "fasta", true); - if (fastafile == "not open") { abort = true; } - else if (fastafile == "not found") { fastafile = ""; m->mothurOut("fasta is a required parameter for the chimera.slayer command."); m->mothurOutEndLine(); abort = true; } + fastafile = validParameter.validFile(parameters, "fasta", false); + if (fastafile == "not found") { fastafile = ""; m->mothurOut("fasta is a required parameter for the chimera.slayer command."); m->mothurOutEndLine(); abort = true; } + else { + splitAtDash(fastafile, fastaFileNames); + + //go through files and make sure they are good, if not, then disregard them + for (int i = 0; i < fastaFileNames.size(); i++) { + if (inputDir != "") { + string path = hasPath(fastaFileNames[i]); + //if the user has not given a path then, add inputdir. else leave path alone. + if (path == "") { fastaFileNames[i] = inputDir + fastaFileNames[i]; } + } + + int ableToOpen; + ifstream in; + + #ifdef USE_MPI + int pid; + MPI_Comm_size(MPI_COMM_WORLD, &processors); //set processors to the number of mpi processes running + MPI_Comm_rank(MPI_COMM_WORLD, &pid); //find out who we are + + if (pid == 0) { + #endif + + ableToOpen = openInputFile(fastaFileNames[i], in, "noerror"); + + //if you can't open it, try default location + if (ableToOpen == 1) { + if (m->getDefaultPath() != "") { //default path is set + string tryPath = m->getDefaultPath() + getSimpleName(fastaFileNames[i]); + m->mothurOut("Unable to open " + fastaFileNames[i] + ". Trying default " + tryPath); m->mothurOutEndLine(); + ableToOpen = openInputFile(tryPath, in, "noerror"); + fastaFileNames[i] = tryPath; + } + } + in.close(); + + #ifdef USE_MPI + for (int j = 1; j < processors; j++) { + MPI_Send(&ableToOpen, 1, MPI_INT, j, 2001, MPI_COMM_WORLD); + } + }else{ + MPI_Status status; + MPI_Recv(&ableToOpen, 1, MPI_INT, 0, 2001, MPI_COMM_WORLD, &status); + } + + #endif + + if (ableToOpen == 1) { + m->mothurOut("Unable to open " + fastaFileNames[i] + ". It will be disregarded."); m->mothurOutEndLine(); + //erase from file list + fastaFileNames.erase(fastaFileNames.begin()+i); + i--; + } + } + + //make sure there is at least one valid file left + if (fastaFileNames.size() == 0) { m->mothurOut("no valid files."); m->mothurOutEndLine(); abort = true; } + } //if the user changes the output directory command factory will send this info to us in the output parameter outputDir = validParameter.validFile(parameters, "outputdir", false); if (outputDir == "not found"){ @@ -106,7 +150,7 @@ ChimeraSlayerCommand::ChimeraSlayerCommand(string option) { temp = validParameter.validFile(parameters, "minbs", false); if (temp == "not found") { temp = "90"; } convert(temp, minBS); - temp = validParameter.validFile(parameters, "minsnp", false); if (temp == "not found") { temp = "10"; } + temp = validParameter.validFile(parameters, "minsnp", false); if (temp == "not found") { temp = "100"; } convert(temp, minSNP); temp = validParameter.validFile(parameters, "parents", false); if (temp == "not found") { temp = "3"; } @@ -141,14 +185,14 @@ void ChimeraSlayerCommand::help(){ m->mothurOut("The chimera.slayer command reads a fastafile and templatefile and outputs potentially chimeric sequences.\n"); m->mothurOut("This command was modeled after the chimeraSlayer written by the Broad Institute.\n"); - m->mothurOut("The chimera.slayer command parameters are fasta, template, filter, mask, processors, ksize, window, match, mismatch, divergence. minsim, mincov, minbs, minsnp, parents, search, iters, increment and numwanted.\n"); //realign, + m->mothurOut("The chimera.slayer command parameters are fasta, template, processors, ksize, window, match, mismatch, divergence. minsim, mincov, minbs, minsnp, parents, search, iters, increment and numwanted.\n"); //realign, m->mothurOut("The fasta parameter allows you to enter the fasta file containing your potentially chimeric sequences, and is required. \n"); + m->mothurOut("You may enter multiple fasta files by separating their names with dashes. ie. fasta=abrecovery.fasta-amazon.fasta \n"); m->mothurOut("The template parameter allows you to enter a template file containing known non-chimeric sequences, and is required. \n"); m->mothurOut("The processors parameter allows you to specify how many processors you would like to use. The default is 1. \n"); #ifdef USE_MPI m->mothurOut("When using MPI, the processors parameter is set to the number of MPI processes running. \n"); #endif - m->mothurOut("The mask parameter allows you to specify a file containing one sequence you wish to use as a mask for the your sequences. \n"); m->mothurOut("The window parameter allows you to specify the window size for searching for chimeras, default=50. \n"); m->mothurOut("The increment parameter allows you to specify how far you move each window while finding chimeric sequences, default=5.\n"); m->mothurOut("The numwanted parameter allows you to specify how many sequences you would each query sequence compared with, default=15.\n"); @@ -161,10 +205,9 @@ void ChimeraSlayerCommand::help(){ m->mothurOut("The minsim parameter allows you to specify a minimum similarity with the parent fragments, default=90. \n"); m->mothurOut("The mincov parameter allows you to specify minimum coverage by closest matches found in template. Default is 70, meaning 70%. \n"); m->mothurOut("The minbs parameter allows you to specify minimum bootstrap support for calling a sequence chimeric. Default is 90, meaning 90%. \n"); - m->mothurOut("The minsnp parameter allows you to specify percent of SNPs to sample on each side of breakpoint for computing bootstrap support (default: 10) \n"); + m->mothurOut("The minsnp parameter allows you to specify percent of SNPs to sample on each side of breakpoint for computing bootstrap support (default: 100) \n"); m->mothurOut("The search parameter allows you to specify search method for finding the closest parent. Choices are distance, blast, and kmer, default distance. \n"); - //m->mothurOut("The realign parameter allows you to realign the query to the potential parents. Choices are true or false, default false. Found to make results worse. \n"); - m->mothurOut("NOT ALL PARAMETERS ARE USED BY ALL METHODS. Please look below for method specifics.\n\n"); + m->mothurOut("The realign parameter allows you to realign the query to the potential parents. Choices are true or false, default false. \n"); m->mothurOut("The chimera.slayer command should be in the following format: \n"); m->mothurOut("chimera.slayer(fasta=yourFastaFile, template=yourTemplate, search=yourSearch) \n"); m->mothurOut("Example: chimera.slayer(fasta=AD.align, template=core_set_aligned.imputed.fasta, search=kmer) \n"); @@ -187,262 +230,241 @@ int ChimeraSlayerCommand::execute(){ if (abort == true) { return 0; } - int start = time(NULL); - - chimera = new ChimeraSlayer(fastafile, templatefile, search, ksize, match, mismatch, window, divR, minSimilarity, minCoverage, minBS, minSNP, parents, iters, increment, numwanted, realign); - - string outputFileName = outputDir + getRootName(getSimpleName(fastafile)) + "slayer.chimeras"; - string accnosFileName = outputDir + getRootName(getSimpleName(fastafile)) + "slayer.accnos"; - bool hasAccnos = true; - - if (m->control_pressed) { delete chimera; return 0; } - - if (chimera->getUnaligned()) { - m->mothurOut("Your template sequences are different lengths, please correct."); m->mothurOutEndLine(); - delete chimera; - return 0; - } - templateSeqsLength = chimera->getLength(); + for (int s = 0; s < fastaFileNames.size(); s++) { + + m->mothurOut("Checking sequences from " + fastaFileNames[s] + " ..." ); m->mothurOutEndLine(); - #ifdef USE_MPI - int pid, end, numSeqsPerProcessor; - int tag = 2001; - vector MPIPos; - MPIWroteAccnos = false; - - MPI_Status status; - MPI_Comm_rank(MPI_COMM_WORLD, &pid); //find out who we are - MPI_Comm_size(MPI_COMM_WORLD, &processors); - - MPI_File inMPI; - MPI_File outMPI; - MPI_File outMPIAccnos; - - int outMode=MPI_MODE_CREATE|MPI_MODE_WRONLY; - int inMode=MPI_MODE_RDONLY; + int start = time(NULL); + + chimera = new ChimeraSlayer(fastaFileNames[s], templatefile, search, ksize, match, mismatch, window, divR, minSimilarity, minCoverage, minBS, minSNP, parents, iters, increment, numwanted, realign); - char outFilename[outputFileName.length()]; - strcpy(outFilename, outputFileName.c_str()); + string outputFileName = outputDir + getRootName(getSimpleName(fastaFileNames[s])) + "slayer.chimeras"; + string accnosFileName = outputDir + getRootName(getSimpleName(fastaFileNames[s])) + "slayer.accnos"; - char outAccnosFilename[accnosFileName.length()]; - strcpy(outAccnosFilename, accnosFileName.c_str()); + if (m->control_pressed) { delete chimera; for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); } return 0; } - char inFileName[fastafile.length()]; - strcpy(inFileName, fastafile.c_str()); - - MPI_File_open(MPI_COMM_WORLD, inFileName, inMode, MPI_INFO_NULL, &inMPI); //comm, filename, mode, info, filepointer - MPI_File_open(MPI_COMM_WORLD, outFilename, outMode, MPI_INFO_NULL, &outMPI); - MPI_File_open(MPI_COMM_WORLD, outAccnosFilename, outMode, MPI_INFO_NULL, &outMPIAccnos); + if (chimera->getUnaligned()) { + m->mothurOut("Your template sequences are different lengths, please correct."); m->mothurOutEndLine(); + delete chimera; + return 0; + } + templateSeqsLength = chimera->getLength(); - if (m->control_pressed) { MPI_File_close(&inMPI); MPI_File_close(&outMPI); MPI_File_close(&outMPIAccnos); delete chimera; return 0; } + #ifdef USE_MPI + int pid, end, numSeqsPerProcessor; + int tag = 2001; + vector MPIPos; + + MPI_Status status; + MPI_Comm_rank(MPI_COMM_WORLD, &pid); //find out who we are + MPI_Comm_size(MPI_COMM_WORLD, &processors); - - if (pid == 0) { //you are the root process - m->mothurOutEndLine(); - m->mothurOut("Only reporting sequence supported by " + toString(minBS) + "% of bootstrapped results."); - m->mothurOutEndLine(); - - string outTemp = "Name\tLeftParent\tRightParent\tDivQLAQRB\tPerIDQLAQRB\tBootStrapA\tDivQLBQRA\tPerIDQLBQRA\tBootStrapB\tFlag\tLeftWindow\tRightWindow\n"; + MPI_File inMPI; + MPI_File outMPI; + MPI_File outMPIAccnos; - //print header - int length = outTemp.length(); - char buf2[length]; - strcpy(buf2, outTemp.c_str()); - MPI_File_write_shared(outMPI, buf2, length, MPI_CHAR, &status); + int outMode=MPI_MODE_CREATE|MPI_MODE_WRONLY; + int inMode=MPI_MODE_RDONLY; - MPIPos = setFilePosFasta(fastafile, numSeqs); //fills MPIPos, returns numSeqs + char outFilename[1024]; + strcpy(outFilename, outputFileName.c_str()); - //send file positions to all processes - MPI_Bcast(&numSeqs, 1, MPI_INT, 0, MPI_COMM_WORLD); //send numSeqs - MPI_Bcast(&MPIPos[0], (numSeqs+1), MPI_LONG, 0, MPI_COMM_WORLD); //send file pos + char outAccnosFilename[1024]; + strcpy(outAccnosFilename, accnosFileName.c_str()); - //figure out how many sequences you have to align - numSeqsPerProcessor = numSeqs / processors; - if(pid == (processors - 1)){ numSeqsPerProcessor = numSeqs - pid * numSeqsPerProcessor; } - int startIndex = pid * numSeqsPerProcessor; + char inFileName[1024]; + strcpy(inFileName, fastaFileNames[s].c_str()); + + MPI_File_open(MPI_COMM_WORLD, inFileName, inMode, MPI_INFO_NULL, &inMPI); //comm, filename, mode, info, filepointer + MPI_File_open(MPI_COMM_WORLD, outFilename, outMode, MPI_INFO_NULL, &outMPI); + MPI_File_open(MPI_COMM_WORLD, outAccnosFilename, outMode, MPI_INFO_NULL, &outMPIAccnos); + + if (m->control_pressed) { MPI_File_close(&inMPI); MPI_File_close(&outMPI); MPI_File_close(&outMPIAccnos); for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); } delete chimera; return 0; } - //align your part - driverMPI(startIndex, numSeqsPerProcessor, inMPI, outMPI, outMPIAccnos, MPIPos); + if (pid == 0) { //you are the root process + m->mothurOutEndLine(); + m->mothurOut("Only reporting sequence supported by " + toString(minBS) + "% of bootstrapped results."); + m->mothurOutEndLine(); + + string outTemp = "Name\tLeftParent\tRightParent\tDivQLAQRB\tPerIDQLAQRB\tBootStrapA\tDivQLBQRA\tPerIDQLBQRA\tBootStrapB\tFlag\tLeftWindow\tRightWindow\n"; + + //print header + int length = outTemp.length(); + char* buf2 = new char[length]; + memcpy(buf2, outTemp.c_str(), length); + + MPI_File_write_shared(outMPI, buf2, length, MPI_CHAR, &status); + delete buf2; + + MPIPos = setFilePosFasta(fastaFileNames[s], numSeqs); //fills MPIPos, returns numSeqs + + //send file positions to all processes + for(int i = 1; i < processors; i++) { + MPI_Send(&numSeqs, 1, MPI_INT, i, tag, MPI_COMM_WORLD); + MPI_Send(&MPIPos[0], (numSeqs+1), MPI_LONG, i, tag, MPI_COMM_WORLD); + } + + //figure out how many sequences you have to align + numSeqsPerProcessor = numSeqs / processors; + int startIndex = pid * numSeqsPerProcessor; + if(pid == (processors - 1)){ numSeqsPerProcessor = numSeqs - pid * numSeqsPerProcessor; } - if (m->control_pressed) { MPI_File_close(&inMPI); MPI_File_close(&outMPI); MPI_File_close(&outMPIAccnos); remove(outputFileName.c_str()); remove(accnosFileName.c_str()); delete chimera; return 0; } + //do your part + driverMPI(startIndex, numSeqsPerProcessor, inMPI, outMPI, outMPIAccnos, MPIPos); + + if (m->control_pressed) { MPI_File_close(&inMPI); MPI_File_close(&outMPI); MPI_File_close(&outMPIAccnos); for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); } remove(outputFileName.c_str()); remove(accnosFileName.c_str()); delete chimera; return 0; } - for (int i = 1; i < processors; i++) { - bool tempResult; - MPI_Recv(&tempResult, 1, MPI_INT, i, tag, MPI_COMM_WORLD, &status); - if (tempResult != 0) { MPIWroteAccnos = true; } + }else{ //you are a child process + MPI_Recv(&numSeqs, 1, MPI_INT, 0, tag, MPI_COMM_WORLD, &status); + MPIPos.resize(numSeqs+1); + MPI_Recv(&MPIPos[0], (numSeqs+1), MPI_LONG, 0, tag, MPI_COMM_WORLD, &status); + + //figure out how many sequences you have to align + numSeqsPerProcessor = numSeqs / processors; + int startIndex = pid * numSeqsPerProcessor; + if(pid == (processors - 1)){ numSeqsPerProcessor = numSeqs - pid * numSeqsPerProcessor; } + + //do your part + driverMPI(startIndex, numSeqsPerProcessor, inMPI, outMPI, outMPIAccnos, MPIPos); + + if (m->control_pressed) { MPI_File_close(&inMPI); MPI_File_close(&outMPI); MPI_File_close(&outMPIAccnos); for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); } delete chimera; return 0; } } - }else{ //you are a child process - MPI_Bcast(&numSeqs, 1, MPI_INT, 0, MPI_COMM_WORLD); //get numSeqs - MPIPos.resize(numSeqs+1); - MPI_Bcast(&MPIPos[0], (numSeqs+1), MPI_LONG, 0, MPI_COMM_WORLD); //get file positions - //figure out how many sequences you have to align - numSeqsPerProcessor = numSeqs / processors; - if(pid == (processors - 1)){ numSeqsPerProcessor = numSeqs - pid * numSeqsPerProcessor; } - int startIndex = pid * numSeqsPerProcessor; + //close files + MPI_File_close(&inMPI); + MPI_File_close(&outMPI); + MPI_File_close(&outMPIAccnos); + MPI_Barrier(MPI_COMM_WORLD); //make everyone wait - just in case - //align your part - driverMPI(startIndex, numSeqsPerProcessor, inMPI, outMPI, outMPIAccnos, MPIPos); + #else + ofstream outHeader; + string tempHeader = outputDir + getRootName(getSimpleName(fastaFileNames[s])) + "slayer.chimeras.tempHeader"; + openOutputFile(tempHeader, outHeader); + + chimera->printHeader(outHeader); + outHeader.close(); + + //break up file + #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) + if(processors == 1){ + ifstream inFASTA; + openInputFile(fastaFileNames[s], inFASTA); + getNumSeqs(inFASTA, numSeqs); + inFASTA.close(); + + lines.push_back(new linePair(0, numSeqs)); + + driver(lines[0], outputFileName, fastaFileNames[s], accnosFileName); + + if (m->control_pressed) { + remove(outputFileName.c_str()); + remove(tempHeader.c_str()); + remove(accnosFileName.c_str()); + for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); } + for (int i = 0; i < lines.size(); i++) { delete lines[i]; } lines.clear(); + delete chimera; + return 0; + } + + }else{ + vector positions; + processIDS.resize(0); + + ifstream inFASTA; + openInputFile(fastaFileNames[s], inFASTA); + + string input; + while(!inFASTA.eof()){ + input = getline(inFASTA); + if (input.length() != 0) { + if(input[0] == '>'){ unsigned long int pos = inFASTA.tellg(); positions.push_back(pos - input.length() - 1); } + } + } + inFASTA.close(); + + numSeqs = positions.size(); + + int numSeqsPerProcessor = numSeqs / processors; + + for (int i = 0; i < processors; i++) { + unsigned long int startPos = positions[ i * numSeqsPerProcessor ]; + if(i == processors - 1){ + numSeqsPerProcessor = numSeqs - i * numSeqsPerProcessor; + } + lines.push_back(new linePair(startPos, numSeqsPerProcessor)); + } + + createProcesses(outputFileName, fastaFileNames[s], accnosFileName); - if (m->control_pressed) { MPI_File_close(&inMPI); MPI_File_close(&outMPI); MPI_File_close(&outMPIAccnos); delete chimera; return 0; } + rename((outputFileName + toString(processIDS[0]) + ".temp").c_str(), outputFileName.c_str()); + rename((accnosFileName + toString(processIDS[0]) + ".temp").c_str(), accnosFileName.c_str()); + + //append output files + for(int i=1;icontrol_pressed) { + remove(outputFileName.c_str()); + remove(accnosFileName.c_str()); + for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); } + for (int i = 0; i < lines.size(); i++) { delete lines[i]; } lines.clear(); + delete chimera; + return 0; + } - MPI_Send(&MPIWroteAccnos, 1, MPI_INT, 0, tag, MPI_COMM_WORLD); - } - - //close files - MPI_File_close(&inMPI); - MPI_File_close(&outMPI); - MPI_File_close(&outMPIAccnos); - - //delete accnos file if blank - if (pid == 0) { - if (!MPIWroteAccnos) { - //MPI_Info info; - //MPI_File_delete(outAccnosFilename, info); - hasAccnos = false; - remove(accnosFileName.c_str()); } - } - - #else - ofstream outHeader; - string tempHeader = outputDir + getRootName(getSimpleName(fastafile)) + "slayer.chimeras.tempHeader"; - openOutputFile(tempHeader, outHeader); - - chimera->printHeader(outHeader); - outHeader.close(); - - //break up file - #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) - if(processors == 1){ + + #else ifstream inFASTA; - openInputFile(fastafile, inFASTA); - numSeqs=count(istreambuf_iterator(inFASTA),istreambuf_iterator(), '>'); + openInputFile(fastaFileNames[s], inFASTA); + getNumSeqs(inFASTA, numSeqs); inFASTA.close(); - lines.push_back(new linePair(0, numSeqs)); - driver(lines[0], outputFileName, fastafile, accnosFileName); + driver(lines[0], outputFileName, fastaFileNames[s], accnosFileName); if (m->control_pressed) { - remove(outputFileName.c_str()); - remove(tempHeader.c_str()); - remove(accnosFileName.c_str()); - for (int i = 0; i < lines.size(); i++) { delete lines[i]; } lines.clear(); - delete chimera; - return 0; - } - - //delete accnos file if its blank - if (isBlank(accnosFileName)) { remove(accnosFileName.c_str()); hasAccnos = false; } - - }else{ - vector positions; - processIDS.resize(0); - - ifstream inFASTA; - openInputFile(fastafile, inFASTA); - - string input; - while(!inFASTA.eof()){ - input = getline(inFASTA); - if (input.length() != 0) { - if(input[0] == '>'){ long int pos = inFASTA.tellg(); positions.push_back(pos - input.length() - 1); } - } - } - inFASTA.close(); - - numSeqs = positions.size(); - - int numSeqsPerProcessor = numSeqs / processors; - - for (int i = 0; i < processors; i++) { - long int startPos = positions[ i * numSeqsPerProcessor ]; - if(i == processors - 1){ - numSeqsPerProcessor = numSeqs - i * numSeqsPerProcessor; - } - lines.push_back(new linePair(startPos, numSeqsPerProcessor)); + remove(outputFileName.c_str()); + remove(tempHeader.c_str()); + remove(accnosFileName.c_str()); + for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); } + for (int i = 0; i < lines.size(); i++) { delete lines[i]; } lines.clear(); + delete chimera; + return 0; } - - createProcesses(outputFileName, fastafile, accnosFileName); + #endif - rename((outputFileName + toString(processIDS[0]) + ".temp").c_str(), outputFileName.c_str()); - - //append output files - for(int i=1;i nonBlankAccnosFiles; - //delete blank accnos files generated with multiple processes - for(int i=0;icontrol_pressed) { - remove(outputFileName.c_str()); - remove(accnosFileName.c_str()); - for (int i = 0; i < lines.size(); i++) { delete lines[i]; } lines.clear(); - delete chimera; - return 0; - } - - } - - #else - ifstream inFASTA; - openInputFile(candidateFileNames[s], inFASTA); - numSeqs=count(istreambuf_iterator(inFASTA),istreambuf_iterator(), '>'); - inFASTA.close(); - lines.push_back(new linePair(0, numSeqs)); - - driver(lines[0], outputFileName, fastafile, accnosFileName); - - if (m->control_pressed) { - remove(outputFileName.c_str()); - remove(tempHeader.c_str()); - remove(accnosFileName.c_str()); - for (int i = 0; i < lines.size(); i++) { delete lines[i]; } lines.clear(); - delete chimera; - return 0; - } + appendFiles(outputFileName, tempHeader); + + remove(outputFileName.c_str()); + rename(tempHeader.c_str(), outputFileName.c_str()); - //delete accnos file if its blank - if (isBlank(accnosFileName)) { remove(accnosFileName.c_str()); hasAccnos = false; } #endif - - appendFiles(tempHeader, outputFileName); - - remove(outputFileName.c_str()); - rename(tempHeader.c_str(), outputFileName.c_str()); - - #endif - delete chimera; + delete chimera; + + + for (int i = 0; i < lines.size(); i++) { delete lines[i]; } lines.clear(); + + outputNames.push_back(outputFileName); + outputNames.push_back(accnosFileName); + + m->mothurOutEndLine(); m->mothurOut("It took " + toString(time(NULL) - start) + " secs to check " + toString(numSeqs) + " sequences."); m->mothurOutEndLine(); + } m->mothurOutEndLine(); m->mothurOut("Output File Names: "); m->mothurOutEndLine(); - m->mothurOut(outputFileName); m->mothurOutEndLine(); - if (hasAccnos) { m->mothurOut(accnosFileName); m->mothurOutEndLine(); } + for (int i = 0; i < outputNames.size(); i++) { m->mothurOut(outputNames[i]); m->mothurOutEndLine(); } m->mothurOutEndLine(); - for (int i = 0; i < lines.size(); i++) { delete lines[i]; } lines.clear(); - - m->mothurOutEndLine(); m->mothurOut("It took " + toString(time(NULL) - start) + " secs to check " + toString(numSeqs) + " sequences."); m->mothurOutEndLine(); - return 0; } @@ -507,9 +529,8 @@ int ChimeraSlayerCommand::driver(linePair* line, string outputFName, string file } //********************************************************************************************************************** #ifdef USE_MPI -int ChimeraSlayerCommand::driverMPI(int start, int num, MPI_File& inMPI, MPI_File& outMPI, MPI_File& outAccMPI, vector& MPIPos){ - try { - +int ChimeraSlayerCommand::driverMPI(int start, int num, MPI_File& inMPI, MPI_File& outMPI, MPI_File& outAccMPI, vector& MPIPos){ + try { MPI_Status status; int pid; MPI_Comm_rank(MPI_COMM_WORLD, &pid); //find out who we are @@ -520,29 +541,31 @@ int ChimeraSlayerCommand::driverMPI(int start, int num, MPI_File& inMPI, MPI_Fil //read next sequence int length = MPIPos[start+i+1] - MPIPos[start+i]; - - char buf4[length]; + + char* buf4 = new char[length]; MPI_File_read_at(inMPI, MPIPos[start+i], buf4, length, MPI_CHAR, &status); - + string tempBuf = buf4; if (tempBuf.length() > length) { tempBuf = tempBuf.substr(0, length); } istringstream iss (tempBuf,istringstream::in); + delete buf4; + Sequence* candidateSeq = new Sequence(iss); gobble(iss); - + if (candidateSeq->getName() != "") { //incase there is a commented sequence at the end of a file if (candidateSeq->getAligned().length() != templateSeqsLength) { m->mothurOut(candidateSeq->getName() + " is not the same length as the template sequences. Skipping."); m->mothurOutEndLine(); }else{ + //find chimeras chimera->getChimeras(candidateSeq); - + if (m->control_pressed) { delete candidateSeq; return 1; } - + //cout << "about to print" << endl; //print results bool isChimeric = chimera->print(outMPI, outAccMPI); - if (isChimeric) { MPIWroteAccnos = true; } } } delete candidateSeq;