X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=chimeraslayercommand.cpp;h=38c097f18b8124f1dadbd657af8413fc74f895c8;hb=4169642e8a8d45f71a4a7241ee02f1b1aae29520;hp=35f3b988f86ec2e868d8eb5a825efc853e1662e7;hpb=b45e6f3d505a212cebad5ca22e2611b1466bca48;p=mothur.git diff --git a/chimeraslayercommand.cpp b/chimeraslayercommand.cpp index 35f3b98..38c097f 100644 --- a/chimeraslayercommand.cpp +++ b/chimeraslayercommand.cpp @@ -9,20 +9,69 @@ #include "chimeraslayercommand.h" #include "chimeraslayer.h" +#include "deconvolutecommand.h" - +//********************************************************************************************************************** +vector ChimeraSlayerCommand::getValidParameters(){ + try { + string AlignArray[] = {"fasta", "processors","trim", "name","window", "include","template","numwanted", "ksize", "match","mismatch", + "divergence", "minsim","mincov","minbs", "minsnp","parents", "iters","outputdir","inputdir", "search","realign" }; + vector myArray (AlignArray, AlignArray+(sizeof(AlignArray)/sizeof(string))); + return myArray; + } + catch(exception& e) { + m->errorOut(e, "ChimeraSlayerCommand", "getValidParameters"); + exit(1); + } +} +//********************************************************************************************************************** +ChimeraSlayerCommand::ChimeraSlayerCommand(){ + try { + abort = true; calledHelp = true; + vector tempOutNames; + outputTypes["chimera"] = tempOutNames; + outputTypes["accnos"] = tempOutNames; + outputTypes["fasta"] = tempOutNames; + } + catch(exception& e) { + m->errorOut(e, "ChimeraSlayerCommand", "ChimeraSlayerCommand"); + exit(1); + } +} +//********************************************************************************************************************** +vector ChimeraSlayerCommand::getRequiredParameters(){ + try { + string AlignArray[] = {"template","fasta"}; + vector myArray (AlignArray, AlignArray+(sizeof(AlignArray)/sizeof(string))); + return myArray; + } + catch(exception& e) { + m->errorOut(e, "ChimeraSlayerCommand", "getRequiredParameters"); + exit(1); + } +} +//********************************************************************************************************************** +vector ChimeraSlayerCommand::getRequiredFiles(){ + try { + vector myArray; + return myArray; + } + catch(exception& e) { + m->errorOut(e, "ChimeraSlayerCommand", "getRequiredFiles"); + exit(1); + } +} //*************************************************************************************************************** - ChimeraSlayerCommand::ChimeraSlayerCommand(string option) { try { - abort = false; + abort = false; calledHelp = false; //allow user to run help - if(option == "help") { help(); abort = true; } + if(option == "help") { help(); abort = true; calledHelp = true; } else { //valid paramters for this command - string Array[] = {"fasta", "processors", "window", "template","numwanted", "ksize", "match","mismatch", + string Array[] = {"fasta", "processors","name", "include","trim", "window", "template","numwanted", "ksize", "match","mismatch", "divergence", "minsim","mincov","minbs", "minsnp","parents", "iters","outputdir","inputdir", "search","realign" }; vector myArray (Array, Array+(sizeof(Array)/sizeof(string))); @@ -37,31 +86,25 @@ ChimeraSlayerCommand::ChimeraSlayerCommand(string option) { if (validParameter.isValidParameter(it->first, myArray, it->second) != true) { abort = true; } } + vector tempOutNames; + outputTypes["chimera"] = tempOutNames; + outputTypes["accnos"] = tempOutNames; + outputTypes["fasta"] = tempOutNames; + //if the user changes the input directory command factory will send this info to us in the output parameter string inputDir = validParameter.validFile(parameters, "inputdir", false); if (inputDir == "not found"){ inputDir = ""; } - else { - string path; - it = parameters.find("template"); - //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["template"] = inputDir + it->second; } - } - } - - + //check for required parameters 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; } + if (fastafile == "not found") { fastafile = ""; m->mothurOut("[ERROR]: fasta is a required parameter for the chimera.slayer command."); m->mothurOutEndLine(); abort = true; } else { - splitAtDash(fastafile, fastaFileNames); + m->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]); + string path = m->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]; } } @@ -69,17 +112,31 @@ ChimeraSlayerCommand::ChimeraSlayerCommand(string option) { int ableToOpen; ifstream in; - ableToOpen = openInputFile(fastaFileNames[i], in, "noerror"); + ableToOpen = m->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]); + string tryPath = m->getDefaultPath() + m->getSimpleName(fastaFileNames[i]); m->mothurOut("Unable to open " + fastaFileNames[i] + ". Trying default " + tryPath); m->mothurOutEndLine(); - ableToOpen = openInputFile(tryPath, in, "noerror"); + ifstream in2; + ableToOpen = m->openInputFile(tryPath, in2, "noerror"); + in2.close(); + fastaFileNames[i] = tryPath; + } + } + + if (ableToOpen == 1) { + if (m->getOutputDir() != "") { //default path is set + string tryPath = m->getOutputDir() + m->getSimpleName(fastaFileNames[i]); + m->mothurOut("Unable to open " + fastaFileNames[i] + ". Trying output directory " + tryPath); m->mothurOutEndLine(); + ifstream in2; + ableToOpen = m->openInputFile(tryPath, in2, "noerror"); + in2.close(); fastaFileNames[i] = tryPath; } } + in.close(); if (ableToOpen == 1) { @@ -91,22 +148,95 @@ ChimeraSlayerCommand::ChimeraSlayerCommand(string option) { } //make sure there is at least one valid file left - if (fastaFileNames.size() == 0) { m->mothurOut("no valid files."); m->mothurOutEndLine(); abort = true; } + if (fastaFileNames.size() == 0) { m->mothurOut("[ERROR]: no valid files."); m->mothurOutEndLine(); abort = true; } } + + //check for required parameters + bool hasName = true; + namefile = validParameter.validFile(parameters, "name", false); + if (namefile == "not found") { namefile = ""; hasName = false; } + else { + m->splitAtDash(namefile, nameFileNames); + + //go through files and make sure they are good, if not, then disregard them + for (int i = 0; i < nameFileNames.size(); i++) { + if (inputDir != "") { + string path = m->hasPath(nameFileNames[i]); + //if the user has not given a path then, add inputdir. else leave path alone. + if (path == "") { nameFileNames[i] = inputDir + nameFileNames[i]; } + } + + int ableToOpen; + ifstream in; + + ableToOpen = m->openInputFile(nameFileNames[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() + m->getSimpleName(nameFileNames[i]); + m->mothurOut("Unable to open " + nameFileNames[i] + ". Trying default " + tryPath); m->mothurOutEndLine(); + ifstream in2; + ableToOpen = m->openInputFile(tryPath, in2, "noerror"); + in2.close(); + nameFileNames[i] = tryPath; + } + } + + if (ableToOpen == 1) { + if (m->getOutputDir() != "") { //default path is set + string tryPath = m->getOutputDir() + m->getSimpleName(nameFileNames[i]); + m->mothurOut("Unable to open " + nameFileNames[i] + ". Trying output directory " + tryPath); m->mothurOutEndLine(); + ifstream in2; + ableToOpen = m->openInputFile(tryPath, in2, "noerror"); + in2.close(); + nameFileNames[i] = tryPath; + } + } + + in.close(); + + if (ableToOpen == 1) { + m->mothurOut("Unable to open " + nameFileNames[i] + ". It will be disregarded."); m->mothurOutEndLine(); + //erase from file list + nameFileNames.erase(nameFileNames.begin()+i); + i--; + } + } + + //make sure there is at least one valid file left + if (nameFileNames.size() == 0) { m->mothurOut("[ERROR]: no valid name files."); m->mothurOutEndLine(); abort = true; } + } + + if (hasName && (nameFileNames.size() != fastaFileNames.size())) { m->mothurOut("[ERROR]: The number of namefiles does not match the number of fastafiles, please correct."); 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"){ - outputDir = ""; - outputDir += hasPath(fastafile); //if user entered a file with a path then preserve it + outputDir = validParameter.validFile(parameters, "outputdir", false); if (outputDir == "not found"){ outputDir = ""; } + + + string path; + it = parameters.find("template"); + //user has given a template file + if(it != parameters.end()){ + if (it->second == "self") { templatefile = "self"; } + else { + path = m->hasPath(it->second); + //if the user has not given a path then, add inputdir. else leave path alone. + if (path == "") { parameters["template"] = inputDir + it->second; } + + templatefile = validParameter.validFile(parameters, "template", true); + if (templatefile == "not open") { abort = true; } + else if (templatefile == "not found") { templatefile = ""; m->mothurOut("template is a required parameter for the chimera.slayer command."); m->mothurOutEndLine(); abort = true; } + } } - - templatefile = validParameter.validFile(parameters, "template", true); - if (templatefile == "not open") { abort = true; } - else if (templatefile == "not found") { templatefile = ""; m->mothurOut("template is a required parameter for the chimera.slayer command."); m->mothurOutEndLine(); abort = true; } - + string temp = validParameter.validFile(parameters, "processors", false); if (temp == "not found") { temp = "1"; } convert(temp, processors); + includeAbunds = validParameter.validFile(parameters, "include", false); if (includeAbunds == "not found") { includeAbunds = "greater"; } + if ((includeAbunds != "greater") && (includeAbunds != "greaterequal") && (includeAbunds != "all")) { includeAbunds = "greater"; m->mothurOut("Invalid include setting. options are greater, greaterequal or all. using greater."); m->mothurOutEndLine(); } + temp = validParameter.validFile(parameters, "ksize", false); if (temp == "not found") { temp = "7"; } convert(temp, ksize); @@ -138,7 +268,13 @@ ChimeraSlayerCommand::ChimeraSlayerCommand(string option) { convert(temp, parents); temp = validParameter.validFile(parameters, "realign", false); if (temp == "not found") { temp = "f"; } - realign = isTrue(temp); + realign = m->isTrue(temp); + + temp = validParameter.validFile(parameters, "trim", false); if (temp == "not found") { temp = "f"; } + trim = m->isTrue(temp); + + //temp = validParameter.validFile(parameters, "trimera", false); if (temp == "not found") { temp = "f"; } + //trimera = m->isTrue(temp); search = validParameter.validFile(parameters, "search", false); if (search == "not found") { search = "distance"; } @@ -166,14 +302,17 @@ 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, 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, name, template, processors, trim, 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("The name parameter allows you to provide a name file, if you are using template=self. \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 template parameter allows you to enter a template file containing known non-chimeric sequences, and is required. You may also set template=self, in this case the abundant sequences will be used as potential parents. \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 trim parameter allows you to output a new fasta file containing your sequences with the chimeric ones trimmed to include only their longest piece, default=F. \n"); + //m->mothurOut("The trimera parameter allows you to check both peices of a chimeric sequence for chimeras, thus looking for trimeras and quadmeras. default=F. \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"); @@ -209,7 +348,7 @@ ChimeraSlayerCommand::~ChimeraSlayerCommand(){ /* do nothing */ } int ChimeraSlayerCommand::execute(){ try{ - if (abort == true) { return 0; } + if (abort == true) { if (calledHelp) { return 0; } return 2; } for (int s = 0; s < fastaFileNames.size(); s++) { @@ -217,10 +356,40 @@ int ChimeraSlayerCommand::execute(){ 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); - - string outputFileName = outputDir + getRootName(getSimpleName(fastaFileNames[s])) + "slayer.chimeras"; - string accnosFileName = outputDir + getRootName(getSimpleName(fastaFileNames[s])) + "slayer.accnos"; + if (templatefile != "self") { //you want to run slayer with a refernce template + chimera = new ChimeraSlayer(fastaFileNames[s], templatefile, trim, search, ksize, match, mismatch, window, divR, minSimilarity, minCoverage, minBS, minSNP, parents, iters, increment, numwanted, realign); + }else { + if (nameFileNames.size() != 0) { //you provided a namefile and we don't need to create one + chimera = new ChimeraSlayer(fastaFileNames[s], templatefile, trim, nameFileNames[s], search, includeAbunds, ksize, match, mismatch, window, divR, minSimilarity, minCoverage, minBS, minSNP, parents, iters, increment, numwanted, realign); + }else { + + m->mothurOutEndLine(); m->mothurOut("No namesfile given, running unique.seqs command to generate one."); m->mothurOutEndLine(); m->mothurOutEndLine(); + + //use unique.seqs to create new name and fastafile + string inputString = "fasta=" + fastaFileNames[s]; + m->mothurOut("/******************************************/"); m->mothurOutEndLine(); + m->mothurOut("Running command: unique.seqs(" + inputString + ")"); m->mothurOutEndLine(); + + Command* uniqueCommand = new DeconvoluteCommand(inputString); + uniqueCommand->execute(); + + map > filenames = uniqueCommand->getOutputFiles(); + + delete uniqueCommand; + + m->mothurOut("/******************************************/"); m->mothurOutEndLine(); + + string nameFile = filenames["name"][0]; + fastaFileNames[s] = filenames["fasta"][0]; + + chimera = new ChimeraSlayer(fastaFileNames[s], templatefile, trim, nameFile, search, includeAbunds, ksize, match, mismatch, window, divR, minSimilarity, minCoverage, minBS, minSNP, parents, iters, increment, numwanted, realign); + } + } + + if (outputDir == "") { outputDir = m->hasPath(fastaFileNames[s]); }//if user entered a file with a path then preserve it + string outputFileName = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + "slayer.chimera"; + string accnosFileName = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + "slayer.accnos"; + string trimFastaFileName = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + "slayer.fasta"; if (m->control_pressed) { delete chimera; for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); } return 0; } @@ -232,7 +401,7 @@ int ChimeraSlayerCommand::execute(){ templateSeqsLength = chimera->getLength(); #ifdef USE_MPI - int pid, end, numSeqsPerProcessor; + int pid, numSeqsPerProcessor; int tag = 2001; vector MPIPos; @@ -243,6 +412,7 @@ int ChimeraSlayerCommand::execute(){ MPI_File inMPI; MPI_File outMPI; MPI_File outMPIAccnos; + MPI_File outMPIFasta; int outMode=MPI_MODE_CREATE|MPI_MODE_WRONLY; int inMode=MPI_MODE_RDONLY; @@ -252,6 +422,9 @@ int ChimeraSlayerCommand::execute(){ char outAccnosFilename[1024]; strcpy(outAccnosFilename, accnosFileName.c_str()); + + char outFastaFilename[1024]; + strcpy(outFastaFilename, trimFastaFileName.c_str()); char inFileName[1024]; strcpy(inFileName, fastaFileNames[s].c_str()); @@ -259,8 +432,9 @@ int ChimeraSlayerCommand::execute(){ 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 (trim) { MPI_File_open(MPI_COMM_WORLD, outFastaFilename, outMode, MPI_INFO_NULL, &outMPIFasta); } - 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; } + if (m->control_pressed) { outputTypes.clear(); MPI_File_close(&inMPI); MPI_File_close(&outMPI); if (trim) { MPI_File_close(&outMPIFasta); } MPI_File_close(&outMPIAccnos); for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); } delete chimera; return 0; } if (pid == 0) { //you are the root process m->mothurOutEndLine(); @@ -277,7 +451,7 @@ int ChimeraSlayerCommand::execute(){ MPI_File_write_shared(outMPI, buf2, length, MPI_CHAR, &status); delete buf2; - MPIPos = setFilePosFasta(fastaFileNames[s], numSeqs); //fills MPIPos, returns numSeqs + MPIPos = m->setFilePosFasta(fastaFileNames[s], numSeqs); //fills MPIPos, returns numSeqs //send file positions to all processes for(int i = 1; i < processors; i++) { @@ -291,9 +465,9 @@ int ChimeraSlayerCommand::execute(){ if(pid == (processors - 1)){ numSeqsPerProcessor = numSeqs - pid * numSeqsPerProcessor; } //do your part - driverMPI(startIndex, numSeqsPerProcessor, inMPI, outMPI, outMPIAccnos, MPIPos); + driverMPI(startIndex, numSeqsPerProcessor, inMPI, outMPI, outMPIAccnos, outMPIFasta, 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; } + if (m->control_pressed) { outputTypes.clear(); MPI_File_close(&inMPI); MPI_File_close(&outMPI); if (trim) { MPI_File_close(&outMPIFasta); } 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; } }else{ //you are a child process MPI_Recv(&numSeqs, 1, MPI_INT, 0, tag, MPI_COMM_WORLD, &status); @@ -306,26 +480,27 @@ int ChimeraSlayerCommand::execute(){ if(pid == (processors - 1)){ numSeqsPerProcessor = numSeqs - pid * numSeqsPerProcessor; } //do your part - driverMPI(startIndex, numSeqsPerProcessor, inMPI, outMPI, outMPIAccnos, MPIPos); + driverMPI(startIndex, numSeqsPerProcessor, inMPI, outMPI, outMPIAccnos, outMPIFasta, 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; } + if (m->control_pressed) { outputTypes.clear(); MPI_File_close(&inMPI); MPI_File_close(&outMPI); if (trim) { MPI_File_close(&outMPIFasta); } MPI_File_close(&outMPIAccnos); for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); } delete chimera; return 0; } } //close files MPI_File_close(&inMPI); MPI_File_close(&outMPI); - MPI_File_close(&outMPIAccnos); + MPI_File_close(&outMPIAccnos); + if (trim) { MPI_File_close(&outMPIFasta); } MPI_Barrier(MPI_COMM_WORLD); //make everyone wait - just in case #else ofstream outHeader; - string tempHeader = outputDir + getRootName(getSimpleName(fastaFileNames[s])) + "slayer.chimeras.tempHeader"; - openOutputFile(tempHeader, outHeader); + string tempHeader = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + "slayer.chimeras.tempHeader"; + m->openOutputFile(tempHeader, outHeader); chimera->printHeader(outHeader); outHeader.close(); - vector positions = divideFile(fastaFileNames[s], processors); + vector positions = m->divideFile(fastaFileNames[s], processors); for (int i = 0; i < (positions.size()-1); i++) { lines.push_back(new linePair(positions[i], positions[(i+1)])); @@ -334,41 +509,49 @@ int ChimeraSlayerCommand::execute(){ //break up file #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) if(processors == 1){ - numSeqs = driver(lines[0], outputFileName, fastaFileNames[s], accnosFileName); + numSeqs = driver(lines[0], outputFileName, fastaFileNames[s], accnosFileName, trimFastaFileName); - 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; } + if (m->control_pressed) { outputTypes.clear(); if (trim) { remove(trimFastaFileName.c_str()); } 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{ processIDS.resize(0); - numSeqs = createProcesses(outputFileName, fastaFileNames[s], accnosFileName); + numSeqs = createProcesses(outputFileName, fastaFileNames[s], accnosFileName, trimFastaFileName); rename((outputFileName + toString(processIDS[0]) + ".temp").c_str(), outputFileName.c_str()); rename((accnosFileName + toString(processIDS[0]) + ".temp").c_str(), accnosFileName.c_str()); + if (trim) { rename((trimFastaFileName + toString(processIDS[0]) + ".temp").c_str(), trimFastaFileName.c_str()); } //append output files for(int i=1;iappendFiles((outputFileName + toString(processIDS[i]) + ".temp"), outputFileName); remove((outputFileName + toString(processIDS[i]) + ".temp").c_str()); } //append output files for(int i=1;iappendFiles((accnosFileName + toString(processIDS[i]) + ".temp"), accnosFileName); remove((accnosFileName + toString(processIDS[i]) + ".temp").c_str()); } - if (m->control_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; } + if (trim) { + for(int i=1;iappendFiles((trimFastaFileName + toString(processIDS[i]) + ".temp"), trimFastaFileName); + remove((trimFastaFileName + toString(processIDS[i]) + ".temp").c_str()); + } + } + + if (m->control_pressed) { outputTypes.clear(); if (trim) { remove(trimFastaFileName.c_str()); } 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; } } #else - numSeqs = driver(lines[0], outputFileName, fastaFileNames[s], accnosFileName); + numSeqs = driver(lines[0], outputFileName, fastaFileNames[s], accnosFileName, trimFastaFileName); - 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; } + if (m->control_pressed) { outputTypes.clear(); if (trim) { remove(trimFastaFileName.c_str()); } 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; } #endif - appendFiles(outputFileName, tempHeader); + m->appendFiles(outputFileName, tempHeader); remove(outputFileName.c_str()); rename(tempHeader.c_str(), outputFileName.c_str()); @@ -379,8 +562,9 @@ int ChimeraSlayerCommand::execute(){ for (int i = 0; i < lines.size(); i++) { delete lines[i]; } lines.clear(); - outputNames.push_back(outputFileName); - outputNames.push_back(accnosFileName); + outputNames.push_back(outputFileName); outputTypes["chimera"].push_back(outputFileName); + outputNames.push_back(accnosFileName); outputTypes["accnos"].push_back(accnosFileName); + if (trim) { outputNames.push_back(trimFastaFileName); outputTypes["fasta"].push_back(trimFastaFileName); } m->mothurOutEndLine(); m->mothurOut("It took " + toString(time(NULL) - start) + " secs to check " + toString(numSeqs) + " sequences."); m->mothurOutEndLine(); } @@ -400,16 +584,19 @@ int ChimeraSlayerCommand::execute(){ } //********************************************************************************************************************** -int ChimeraSlayerCommand::driver(linePair* filePos, string outputFName, string filename, string accnos){ +int ChimeraSlayerCommand::driver(linePair* filePos, string outputFName, string filename, string accnos, string fasta){ try { ofstream out; - openOutputFile(outputFName, out); + m->openOutputFile(outputFName, out); ofstream out2; - openOutputFile(accnos, out2); + m->openOutputFile(accnos, out2); + + ofstream out3; + if (trim) { m->openOutputFile(fasta, out3); } ifstream inFASTA; - openInputFile(filename, inFASTA); + m->openInputFile(filename, inFASTA); inFASTA.seekg(filePos->start); @@ -418,9 +605,9 @@ int ChimeraSlayerCommand::driver(linePair* filePos, string outputFName, string f while (!done) { - if (m->control_pressed) { return 1; } + if (m->control_pressed) { out.close(); out2.close(); if (trim) { out3.close(); } inFASTA.close(); return 1; } - Sequence* candidateSeq = new Sequence(inFASTA); gobble(inFASTA); + Sequence* candidateSeq = new Sequence(inFASTA); m->gobble(inFASTA); if (candidateSeq->getName() != "") { //incase there is a commented sequence at the end of a file @@ -433,14 +620,23 @@ int ChimeraSlayerCommand::driver(linePair* filePos, string outputFName, string f if (m->control_pressed) { delete candidateSeq; return 1; } //print results - chimera->print(out, out2); + Sequence* trimmed = chimera->print(out, out2); + + if (trim) { trimmed->printSequence(out3); delete trimmed; } + + //do you want to check both pieces for chimeras + //if (trimera) {} } count++; } delete candidateSeq; - unsigned long int pos = inFASTA.tellg(); - if ((pos == -1) || (pos >= filePos->end)) { break; } + #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) + unsigned long int pos = inFASTA.tellg(); + if ((pos == -1) || (pos >= filePos->end)) { break; } + #else + if (inFASTA.eof()) { break; } + #endif //report progress if((count) % 100 == 0){ m->mothurOut("Processing sequence: " + toString(count)); m->mothurOutEndLine(); } @@ -450,6 +646,7 @@ int ChimeraSlayerCommand::driver(linePair* filePos, string outputFName, string f out.close(); out2.close(); + if (trim) { out3.close(); } inFASTA.close(); return count; @@ -461,7 +658,7 @@ int ChimeraSlayerCommand::driver(linePair* filePos, string outputFName, string f } //********************************************************************************************************************** #ifdef USE_MPI -int ChimeraSlayerCommand::driverMPI(int start, int num, MPI_File& inMPI, MPI_File& outMPI, MPI_File& outAccMPI, vector& MPIPos){ +int ChimeraSlayerCommand::driverMPI(int start, int num, MPI_File& inMPI, MPI_File& outMPI, MPI_File& outAccMPI, MPI_File& outFastaMPI, vector& MPIPos){ try { MPI_Status status; int pid; @@ -483,7 +680,7 @@ int ChimeraSlayerCommand::driverMPI(int start, int num, MPI_File& inMPI, MPI_Fil delete buf4; - Sequence* candidateSeq = new Sequence(iss); gobble(iss); + Sequence* candidateSeq = new Sequence(iss); m->gobble(iss); if (candidateSeq->getName() != "") { //incase there is a commented sequence at the end of a file @@ -495,9 +692,26 @@ int ChimeraSlayerCommand::driverMPI(int start, int num, MPI_File& inMPI, MPI_Fil chimera->getChimeras(candidateSeq); if (m->control_pressed) { delete candidateSeq; return 1; } - //cout << "about to print" << endl; + //print results - bool isChimeric = chimera->print(outMPI, outAccMPI); + Sequence* trimmed = chimera->print(outMPI, outAccMPI); + + if (trim) { + string outputString = ">" + trimmed->getName() + "\n" + trimmed->getAligned() + "\n"; + delete trimmed; + + //write to accnos file + int length = outputString.length(); + char* buf2 = new char[length]; + memcpy(buf2, outputString.c_str(), length); + + MPI_File_write_shared(outFastaMPI, buf2, length, MPI_CHAR, &status); + delete buf2; + } + + //do you want to check both pieces for chimeras + //if (trimera) {} + } } delete candidateSeq; @@ -520,7 +734,7 @@ int ChimeraSlayerCommand::driverMPI(int start, int num, MPI_File& inMPI, MPI_Fil /**************************************************************************************************/ -int ChimeraSlayerCommand::createProcesses(string outputFileName, string filename, string accnos) { +int ChimeraSlayerCommand::createProcesses(string outputFileName, string filename, string accnos, string fasta) { try { #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) int process = 0; @@ -534,17 +748,21 @@ int ChimeraSlayerCommand::createProcesses(string outputFileName, string filename processIDS.push_back(pid); //create map from line number to pid so you can append files in correct order later process++; }else if (pid == 0){ - num = driver(lines[process], outputFileName + toString(getpid()) + ".temp", filename, accnos + toString(getpid()) + ".temp"); + num = driver(lines[process], outputFileName + toString(getpid()) + ".temp", filename, accnos + toString(getpid()) + ".temp", fasta + toString(getpid()) + ".temp"); //pass numSeqs to parent ofstream out; string tempFile = outputFileName + toString(getpid()) + ".num.temp"; - openOutputFile(tempFile, out); + m->openOutputFile(tempFile, out); out << num << endl; out.close(); exit(0); - }else { m->mothurOut("unable to spawn the necessary processes."); m->mothurOutEndLine(); exit(0); } + }else { + m->mothurOut("[ERROR]: unable to spawn the necessary processes."); m->mothurOutEndLine(); + for (int i = 0; i < processIDS.size(); i++) { kill (processIDS[i], SIGINT); } + exit(0); + } } //force parent to wait until all the processes are done @@ -556,7 +774,7 @@ int ChimeraSlayerCommand::createProcesses(string outputFileName, string filename for (int i = 0; i < processIDS.size(); i++) { ifstream in; string tempFile = outputFileName + toString(processIDS[i]) + ".num.temp"; - openInputFile(tempFile, in); + m->openInputFile(tempFile, in); if (!in.eof()) { int tempNum = 0; in >> tempNum; num += tempNum; } in.close(); remove(tempFile.c_str()); }