X-Git-Url: https://git.donarmstrong.com/?p=mothur.git;a=blobdiff_plain;f=aligncommand.cpp;h=f757a7920bb337883a2a7773f35ed533d2186c80;hp=87ceb2c41a2f22e8a3b55a672641a06095f7f0f2;hb=050a3ff02473a3d4c0980964e1a9ebe52e55d6b8;hpb=990ded2eb38078a417fa00b4fbb4d3c24f4b5046 diff --git a/aligncommand.cpp b/aligncommand.cpp index 87ceb2c..f757a79 100644 --- a/aligncommand.cpp +++ b/aligncommand.cpp @@ -20,21 +20,21 @@ //********************************************************************************************************************** vector AlignCommand::setParameters(){ try { - CommandParameter ptemplate("reference", "InputTypes", "", "", "none", "none", "none",false,true); parameters.push_back(ptemplate); - CommandParameter pcandidate("fasta", "InputTypes", "", "", "none", "none", "none",false,true); parameters.push_back(pcandidate); - CommandParameter psearch("search", "Multiple", "kmer-blast-suffix", "kmer", "", "", "",false,false); parameters.push_back(psearch); - CommandParameter pksize("ksize", "Number", "", "8", "", "", "",false,false); parameters.push_back(pksize); - CommandParameter pmatch("match", "Number", "", "1.0", "", "", "",false,false); parameters.push_back(pmatch); - CommandParameter palign("align", "Multiple", "needleman-gotoh-blast-noalign", "needleman", "", "", "",false,false); parameters.push_back(palign); - CommandParameter pmismatch("mismatch", "Number", "", "-1.0", "", "", "",false,false); parameters.push_back(pmismatch); - CommandParameter pgapopen("gapopen", "Number", "", "-2.0", "", "", "",false,false); parameters.push_back(pgapopen); - CommandParameter pgapextend("gapextend", "Number", "", "-1.0", "", "", "",false,false); parameters.push_back(pgapextend); - CommandParameter pprocessors("processors", "Number", "", "1", "", "", "",false,false); parameters.push_back(pprocessors); - CommandParameter pflip("flip", "Boolean", "", "F", "", "", "",false,false); parameters.push_back(pflip); - CommandParameter psave("save", "Boolean", "", "F", "", "", "",false,false); parameters.push_back(psave); - CommandParameter pthreshold("threshold", "Number", "", "0.50", "", "", "",false,false); parameters.push_back(pthreshold); - CommandParameter pinputdir("inputdir", "String", "", "", "", "", "",false,false); parameters.push_back(pinputdir); - CommandParameter poutputdir("outputdir", "String", "", "", "", "", "",false,false); parameters.push_back(poutputdir); + CommandParameter ptemplate("reference", "InputTypes", "", "", "none", "none", "none","",false,true,true); parameters.push_back(ptemplate); + CommandParameter pcandidate("fasta", "InputTypes", "", "", "none", "none", "none","fasta-alignreport-accnos",false,true,true); parameters.push_back(pcandidate); + CommandParameter psearch("search", "Multiple", "kmer-blast-suffix", "kmer", "", "", "","",false,false,true); parameters.push_back(psearch); + CommandParameter pksize("ksize", "Number", "", "8", "", "", "","",false,false); parameters.push_back(pksize); + CommandParameter pmatch("match", "Number", "", "1.0", "", "", "","",false,false); parameters.push_back(pmatch); + CommandParameter palign("align", "Multiple", "needleman-gotoh-blast-noalign", "needleman", "", "", "","",false,false,true); parameters.push_back(palign); + CommandParameter pmismatch("mismatch", "Number", "", "-1.0", "", "", "","",false,false); parameters.push_back(pmismatch); + CommandParameter pgapopen("gapopen", "Number", "", "-5.0", "", "", "","",false,false); parameters.push_back(pgapopen); + CommandParameter pgapextend("gapextend", "Number", "", "-2.0", "", "", "","",false,false); parameters.push_back(pgapextend); + CommandParameter pprocessors("processors", "Number", "", "1", "", "", "","",false,false,true); parameters.push_back(pprocessors); + CommandParameter pflip("flip", "Boolean", "", "F", "", "", "","",false,false); parameters.push_back(pflip); + CommandParameter psave("save", "Boolean", "", "F", "", "", "","",false,false); parameters.push_back(psave); + CommandParameter pthreshold("threshold", "Number", "", "0.50", "", "", "","",false,false); parameters.push_back(pthreshold); + CommandParameter pinputdir("inputdir", "String", "", "", "", "", "","",false,false); parameters.push_back(pinputdir); + CommandParameter poutputdir("outputdir", "String", "", "", "", "", "","",false,false); parameters.push_back(poutputdir); vector myArray; for (int i = 0; i < parameters.size(); i++) { myArray.push_back(parameters[i].name); } @@ -57,8 +57,8 @@ string AlignCommand::getHelpString(){ helpString += "The ksize parameter allows you to specify the kmer size for finding most similar template to candidate. The default is 8."; helpString += "The match parameter allows you to specify the bonus for having the same base. The default is 1.0."; helpString += "The mistmatch parameter allows you to specify the penalty for having different bases. The default is -1.0."; - helpString += "The gapopen parameter allows you to specify the penalty for opening a gap in an alignment. The default is -2.0."; - helpString += "The gapextend parameter allows you to specify the penalty for extending a gap in an alignment. The default is -1.0."; + helpString += "The gapopen parameter allows you to specify the penalty for opening a gap in an alignment. The default is -5.0."; + helpString += "The gapextend parameter allows you to specify the penalty for extending a gap in an alignment. The default is -2.0."; helpString += "The flip parameter is used to specify whether or not you want mothur to try the reverse complement if a sequence falls below the threshold. The default is false."; helpString += "The threshold is used to specify a cutoff at which an alignment is deemed 'bad' and the reverse complement may be tried. The default threshold is 0.50, meaning 50% of the bases are removed in the alignment."; helpString += "If the flip parameter is set to true the reverse complement of the sequence is aligned and the better alignment is reported."; @@ -76,6 +76,23 @@ string AlignCommand::getHelpString(){ } } //********************************************************************************************************************** +string AlignCommand::getOutputPattern(string type) { + try { + string pattern = ""; + + if (type == "fasta") { pattern = "[filename],align"; } //makes file like: amazon.align + else if (type == "alignreport") { pattern = "[filename],align.report"; } + else if (type == "accnos") { pattern = "[filename],flip.accnos"; } + else { m->mothurOut("[ERROR]: No definition for type " + type + " output pattern.\n"); m->control_pressed = true; } + + return pattern; + } + catch(exception& e) { + m->errorOut(e, "AlignCommand", "getOutputPattern"); + exit(1); + } +} +//********************************************************************************************************************** AlignCommand::AlignCommand(){ try { abort = true; calledHelp = true; @@ -224,34 +241,29 @@ AlignCommand::AlignCommand(string option) { // ...at some point should added some additional type checking... string temp; temp = validParameter.validFile(parameters, "ksize", false); if (temp == "not found"){ temp = "8"; } - convert(temp, kmerSize); + m->mothurConvert(temp, kmerSize); temp = validParameter.validFile(parameters, "match", false); if (temp == "not found"){ temp = "1.0"; } - convert(temp, match); + m->mothurConvert(temp, match); temp = validParameter.validFile(parameters, "mismatch", false); if (temp == "not found"){ temp = "-1.0"; } - convert(temp, misMatch); + m->mothurConvert(temp, misMatch); - temp = validParameter.validFile(parameters, "gapopen", false); if (temp == "not found"){ temp = "-2.0"; } - convert(temp, gapOpen); + temp = validParameter.validFile(parameters, "gapopen", false); if (temp == "not found"){ temp = "-5.0"; } + m->mothurConvert(temp, gapOpen); - temp = validParameter.validFile(parameters, "gapextend", false); if (temp == "not found"){ temp = "-1.0"; } - convert(temp, gapExtend); + temp = validParameter.validFile(parameters, "gapextend", false); if (temp == "not found"){ temp = "-2.0"; } + m->mothurConvert(temp, gapExtend); temp = validParameter.validFile(parameters, "processors", false); if (temp == "not found"){ temp = m->getProcessors(); } m->setProcessors(temp); - convert(temp, processors); + m->mothurConvert(temp, processors); temp = validParameter.validFile(parameters, "flip", false); if (temp == "not found"){ temp = "f"; } flip = m->isTrue(temp); temp = validParameter.validFile(parameters, "save", false); if (temp == "not found"){ temp = "f"; } save = m->isTrue(temp); - //this is so the threads can quickly load the reference data - #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) - #else - if (processors != 1) { save = true; } - #endif rdb->save = save; if (save) { //clear out old references rdb->clearMemory(); @@ -272,7 +284,7 @@ AlignCommand::AlignCommand(string option) { else { if (save) { rdb->setSavedReference(templateFileName); } } temp = validParameter.validFile(parameters, "threshold", false); if (temp == "not found"){ temp = "0.50"; } - convert(temp, threshold); + m->mothurConvert(temp, threshold); search = validParameter.validFile(parameters, "search", false); if (search == "not found"){ search = "kmer"; } if ((search != "suffix") && (search != "kmer") && (search != "blast")) { m->mothurOut("invalid search option: choices are kmer, suffix or blast."); m->mothurOutEndLine(); abort=true; } @@ -302,7 +314,7 @@ int AlignCommand::execute(){ try { if (abort == true) { if (calledHelp) { return 0; } return 2; } - templateDB = new AlignmentDB(templateFileName, search, kmerSize, gapOpen, gapExtend, match, misMatch); + templateDB = new AlignmentDB(templateFileName, search, kmerSize, gapOpen, gapExtend, match, misMatch, rand()); for (int s = 0; s < candidateFileNames.size(); s++) { if (m->control_pressed) { outputTypes.clear(); return 0; } @@ -310,9 +322,11 @@ int AlignCommand::execute(){ m->mothurOut("Aligning sequences from " + candidateFileNames[s] + " ..." ); m->mothurOutEndLine(); if (outputDir == "") { outputDir += m->hasPath(candidateFileNames[s]); } - string alignFileName = outputDir + m->getRootName(m->getSimpleName(candidateFileNames[s])) + "align"; - string reportFileName = outputDir + m->getRootName(m->getSimpleName(candidateFileNames[s])) + "align.report"; - string accnosFileName = outputDir + m->getRootName(m->getSimpleName(candidateFileNames[s])) + "flip.accnos"; + map variables; variables["[filename]"] = outputDir + m->getRootName(m->getSimpleName(candidateFileNames[s])); + string alignFileName = getOutputFileName("fasta", variables); + string reportFileName = getOutputFileName("alignreport", variables); + string accnosFileName = getOutputFileName("accnos", variables); + bool hasAccnos = true; int numFastaSeqs = 0; @@ -322,7 +336,7 @@ int AlignCommand::execute(){ #ifdef USE_MPI int pid, numSeqsPerProcessor; int tag = 2001; - vector MPIPos; + vector MPIPos; MPIWroteAccnos = false; MPI_Status status; @@ -426,19 +440,24 @@ int AlignCommand::execute(){ #else - vector positions; - #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) + vector positions; + #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) || (__linux__) || (__unix__) || (__unix) positions = m->divideFile(candidateFileNames[s], processors); for (int i = 0; i < (positions.size()-1); i++) { lines.push_back(new linePair(positions[i], positions[(i+1)])); } #else - positions = m->setFilePosFasta(candidateFileNames[s], numFastaSeqs); - - //figure out how many sequences you have to process - int numSeqsPerProcessor = numFastaSeqs / processors; - for (int i = 0; i < processors; i++) { - int startIndex = i * numSeqsPerProcessor; - if(i == (processors - 1)){ numSeqsPerProcessor = numFastaSeqs - i * numSeqsPerProcessor; } - lines.push_back(new linePair(positions[startIndex], numSeqsPerProcessor)); + if (processors == 1) { + lines.push_back(new linePair(0, 1000)); + }else { + positions = m->setFilePosFasta(candidateFileNames[s], numFastaSeqs); + if (positions.size() < processors) { processors = positions.size(); } + + //figure out how many sequences you have to process + int numSeqsPerProcessor = numFastaSeqs / processors; + for (int i = 0; i < processors; i++) { + int startIndex = i * numSeqsPerProcessor; + if(i == (processors - 1)){ numSeqsPerProcessor = numFastaSeqs - i * numSeqsPerProcessor; } + lines.push_back(new linePair(positions[startIndex], numSeqsPerProcessor)); + } } #endif @@ -539,6 +558,7 @@ int AlignCommand::driver(linePair* filePos, string alignFName, string reportFNam if (m->control_pressed) { break; } Sequence* candidateSeq = new Sequence(inFASTA); m->gobble(inFASTA); + cout << candidateSeq->getAligned() << endl; report.setCandidate(candidateSeq); int origNumBases = candidateSeq->getNumBases(); @@ -549,7 +569,6 @@ int AlignCommand::driver(linePair* filePos, string alignFName, string reportFNam if (candidateSeq->getUnaligned().length() > alignment->getnRows()) { alignment->resize(candidateSeq->getUnaligned().length()+1); } - Sequence temp = templateDB->findClosestSequence(candidateSeq); Sequence* templateSeq = &temp; @@ -618,19 +637,19 @@ int AlignCommand::driver(linePair* filePos, string alignFName, string reportFNam } delete candidateSeq; - #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) - unsigned long int pos = inFASTA.tellg(); + #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) || (__linux__) || (__unix__) || (__unix) + unsigned long long pos = inFASTA.tellg(); if ((pos == -1) || (pos >= filePos->end)) { break; } #else if (inFASTA.eof()) { break; } #endif //report progress - if((count) % 100 == 0){ m->mothurOut(toString(count)); m->mothurOutEndLine(); } + if((count) % 100 == 0){ m->mothurOutJustToScreen(toString(count) + "\n"); } } //report progress - if((count) % 100 != 0){ m->mothurOut(toString(count)); m->mothurOutEndLine(); } + if((count) % 100 != 0){ m->mothurOutJustToScreen(toString(count) + "\n"); } delete alignment; alignmentFile.close(); @@ -646,7 +665,7 @@ int AlignCommand::driver(linePair* filePos, string alignFName, string reportFNam } //********************************************************************************************************************** #ifdef USE_MPI -int AlignCommand::driverMPI(int start, int num, MPI_File& inMPI, MPI_File& alignFile, MPI_File& reportFile, MPI_File& accnosFile, vector& MPIPos){ +int AlignCommand::driverMPI(int start, int num, MPI_File& inMPI, MPI_File& alignFile, MPI_File& reportFile, MPI_File& accnosFile, vector& MPIPos){ try { string outputString = ""; MPI_Status statusReport; @@ -827,7 +846,7 @@ int AlignCommand::createProcesses(string alignFileName, string reportFileName, s try { int num = 0; processIDS.resize(0); -#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) +#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) || (__linux__) || (__unix__) || (__unix) int process = 1; //loop through and create all the processes you want @@ -875,7 +894,7 @@ int AlignCommand::createProcesses(string alignFileName, string reportFileName, s if (!in.eof()) { int tempNum = 0; in >> tempNum; num += tempNum; } in.close(); m->mothurRemove(tempFile); - appendAlignFiles((alignFileName + toString(processIDS[i]) + ".temp"), alignFileName); + m->appendFiles((alignFileName + toString(processIDS[i]) + ".temp"), alignFileName); m->mothurRemove((alignFileName + toString(processIDS[i]) + ".temp")); appendReportFiles((reportFileName + toString(processIDS[i]) + ".temp"), reportFileName); @@ -892,7 +911,7 @@ int AlignCommand::createProcesses(string alignFileName, string reportFileName, s rename(nonBlankAccnosFiles[0].c_str(), accnosFName.c_str()); for (int h=1; h < nonBlankAccnosFiles.size(); h++) { - appendAlignFiles(nonBlankAccnosFiles[h], accnosFName); + m->appendFiles(nonBlankAccnosFiles[h], accnosFName); m->mothurRemove(nonBlankAccnosFiles[h]); } }else { //recreate the accnosfile if needed @@ -919,7 +938,7 @@ int AlignCommand::createProcesses(string alignFileName, string reportFileName, s string extension = ""; if (i != 0) { extension = toString(i) + ".temp"; } - alignData* tempalign = new alignData((alignFileName + extension), (reportFileName + extension), (accnosFName + extension), filename, align, search, kmerSize, m, lines[i]->start, lines[i]->end, flip, match, misMatch, gapOpen, gapExtend, threshold); + alignData* tempalign = new alignData(templateFileName, (alignFileName + extension), (reportFileName + extension), (accnosFName + extension), filename, align, search, kmerSize, m, lines[i]->start, lines[i]->end, flip, match, misMatch, gapOpen, gapExtend, threshold, i); pDataArray.push_back(tempalign); processIDS.push_back(i); @@ -947,6 +966,9 @@ int AlignCommand::createProcesses(string alignFileName, string reportFileName, s //Close all thread handles and free memory allocations. for(int i=0; i < pDataArray.size(); i++){ + if (pDataArray[i]->count != pDataArray[i]->end) { + m->mothurOut("[ERROR]: process " + toString(i) + " only processed " + toString(pDataArray[i]->count) + " of " + toString(pDataArray[i]->end) + " sequences assigned to it, quitting. \n"); m->control_pressed = true; + } num += pDataArray[i]->count; CloseHandle(hThreadArray[i]); delete pDataArray[i]; @@ -957,7 +979,7 @@ int AlignCommand::createProcesses(string alignFileName, string reportFileName, s else { m->mothurRemove(accnosFName); } //remove so other files can be renamed to it for (int i = 1; i < processors; i++) { - appendAlignFiles((alignFileName + toString(i) + ".temp"), alignFileName); + m->appendFiles((alignFileName + toString(i) + ".temp"), alignFileName); m->mothurRemove((alignFileName + toString(i) + ".temp")); appendReportFiles((reportFileName + toString(i) + ".temp"), reportFileName); @@ -973,7 +995,7 @@ int AlignCommand::createProcesses(string alignFileName, string reportFileName, s rename(nonBlankAccnosFiles[0].c_str(), accnosFName.c_str()); for (int h=1; h < nonBlankAccnosFiles.size(); h++) { - appendAlignFiles(nonBlankAccnosFiles[h], accnosFName); + m->appendFiles(nonBlankAccnosFiles[h], accnosFName); m->mothurRemove(nonBlankAccnosFiles[h]); } }else { //recreate the accnosfile if needed @@ -990,29 +1012,6 @@ int AlignCommand::createProcesses(string alignFileName, string reportFileName, s exit(1); } } -/**************************************************************************************************/ - -void AlignCommand::appendAlignFiles(string temp, string filename) { - try{ - - ofstream output; - ifstream input; - m->openOutputFileAppend(filename, output); - m->openInputFile(temp, input); - - while(char c = input.get()){ - if(input.eof()) { break; } - else { output << c; } - } - - input.close(); - output.close(); - } - catch(exception& e) { - m->errorOut(e, "AlignCommand", "appendAlignFiles"); - exit(1); - } -} //********************************************************************************************************************** void AlignCommand::appendReportFiles(string temp, string filename) { @@ -1025,10 +1024,11 @@ void AlignCommand::appendReportFiles(string temp, string filename) { while (!input.eof()) { char c = input.get(); if (c == 10 || c == 13){ break; } } // get header line - while(char c = input.get()){ - if(input.eof()) { break; } - else { output << c; } - } + char buffer[4096]; + while (!input.eof()) { + input.read(buffer, 4096); + output.write(buffer, input.gcount()); + } input.close(); output.close();