X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=aligncommand.cpp;h=3b4b6c93e232a25bab87a855c6955d177926e823;hb=74844a60d80c6dd06e3fb02ee9b928424f9019b0;hp=2811f1fd7e6d404c557dff149cce8860f80f99f0;hpb=cd37904452dc95b183ff313ff05720c562902487;p=mothur.git diff --git a/aligncommand.cpp b/aligncommand.cpp index 2811f1f..3b4b6c9 100644 --- a/aligncommand.cpp +++ b/aligncommand.cpp @@ -22,120 +22,173 @@ #include "blastalign.hpp" #include "noalign.hpp" -#include "kmerdb.hpp" -#include "suffixdb.hpp" -#include "blastdb.hpp" - #include "nast.hpp" #include "nastreport.hpp" //********************************************************************************************************************** -AlignCommand::AlignCommand(string option){ +AlignCommand::AlignCommand(string option) { try { -// globaldata = GlobalData::getInstance(); + abort = false; //allow user to run help if(option == "help") { help(); abort = true; } else { - + //valid paramters for this command - string AlignArray[] = {"template","candidate","search","ksize","align","match","mismatch","gapopen","gapextend", "processors"}; + string AlignArray[] = {"template","candidate","search","ksize","align","match","mismatch","gapopen","gapextend", "processors","flip","threshold","outputdir","inputdir"}; vector myArray (AlignArray, AlignArray+(sizeof(AlignArray)/sizeof(string))); OptionParser parser(option); map parameters = parser.getParameters(); ValidParameters validParameter; - + map::iterator it; + //check to make sure all parameters are valid for command - for (map::iterator it = parameters.begin(); it != parameters.end(); it++) { + for (it = parameters.begin(); it != parameters.end(); it++) { if (validParameter.isValidParameter(it->first, myArray, it->second) != true) { 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 = ""; } + + + //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 templateFileName = validParameter.validFile(parameters, "template", true); - if (templateFileName == "not found") { cout << "template is a required parameter for the align.seqs command." << endl; abort = true; } - else if (templateFileName == "not open") { abort = true; } - - candidateFileName = validParameter.validFile(parameters, "candidate", true); - if (candidateFileName == "not found") { cout << "candidate is a required parameter for the align.seqs command." << endl; abort = true; } - else if (candidateFileName == "not open") { abort = true; } + + if (templateFileName == "not found") { + m->mothurOut("template is a required parameter for the align.seqs command."); + m->mothurOutEndLine(); + abort = true; + }else if (templateFileName == "not open") { abort = true; } + + candidateFileName = validParameter.validFile(parameters, "candidate", false); + if (candidateFileName == "not found") { m->mothurOut("candidate is a required parameter for the align.seqs command."); m->mothurOutEndLine(); abort = true; } + else { + splitAtDash(candidateFileName, candidateFileNames); + + //go through files and make sure they are good, if not, then disregard them + for (int i = 0; i < candidateFileNames.size(); i++) { + if (inputDir != "") { + string path = hasPath(candidateFileNames[i]); + //if the user has not given a path then, add inputdir. else leave path alone. + if (path == "") { candidateFileNames[i] = inputDir + candidateFileNames[i]; } + } + + int ableToOpen; + ifstream in; + ableToOpen = openInputFile(candidateFileNames[i], in); + if (ableToOpen == 1) { + m->mothurOut(candidateFileNames[i] + " will be disregarded."); m->mothurOutEndLine(); + //erase from file list + candidateFileNames.erase(candidateFileNames.begin()+i); + i--; + } + in.close(); + } + + //make sure there is at least one valid file left + if (candidateFileNames.size() == 0) { m->mothurOut("no valid files."); m->mothurOutEndLine(); abort = true; } + } //check for optional parameter and set defaults // ...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); - + temp = validParameter.validFile(parameters, "match", false); if (temp == "not found"){ temp = "1.0"; } convert(temp, match); - + temp = validParameter.validFile(parameters, "mismatch", false); if (temp == "not found"){ temp = "-1.0"; } convert(temp, misMatch); - - temp = validParameter.validFile(parameters, "gapopen", false); if (temp == "not found"){ temp = "-1.0"; } + + temp = validParameter.validFile(parameters, "gapopen", false); if (temp == "not found"){ temp = "-2.0"; } convert(temp, gapOpen); - - temp = validParameter.validFile(parameters, "gapextend", false); if (temp == "not found"){ temp = "-2.0"; } + + temp = validParameter.validFile(parameters, "gapextend", false); if (temp == "not found"){ temp = "-1.0"; } convert(temp, gapExtend); - + temp = validParameter.validFile(parameters, "processors", false); if (temp == "not found"){ temp = "1"; } convert(temp, processors); - + + temp = validParameter.validFile(parameters, "flip", false); if (temp == "not found"){ temp = "f"; } + flip = isTrue(temp); + + temp = validParameter.validFile(parameters, "threshold", false); if (temp == "not found"){ temp = "0.50"; } + convert(temp, threshold); + search = validParameter.validFile(parameters, "search", false); if (search == "not found"){ search = "kmer"; } align = validParameter.validFile(parameters, "align", false); if (align == "not found"){ align = "needleman"; } } - + } catch(exception& e) { - cout << "Standard Error: " << e.what() << " has occurred in the AlignCommand class Function AlignCommand. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + m->errorOut(e, "AlignCommand", "AlignCommand"); exit(1); } - catch(...) { - cout << "An unknown error has occurred in the AlignCommand class function AlignCommand. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; - exit(1); - } } //********************************************************************************************************************** -AlignCommand::~AlignCommand(){ - delete templateDB; - delete alignment; +AlignCommand::~AlignCommand(){ + + if (abort == false) { + for (int i = 0; i < lines.size(); i++) { delete lines[i]; } lines.clear(); + delete templateDB; + delete alignment; + } } //********************************************************************************************************************** void AlignCommand::help(){ try { - cout << "The align.seqs command reads a file containing sequences and creates an alignment file and a report file." << "\n"; - cout << "The align.seqs command parameters are fasta, candidate, search, ksize, align, match, mismatch, gapopen and gapextend. " << "\n"; - cout << "The fasta and candidate parameters are required." << "\n"; - cout << "The search parameter allows you to specify the method to find most similar template. Your options are: suffix, kmer and blast. The default is kmer." << "\n"; - cout << "The align parameter allows you to specify the alignment method to use. Your options are: gotoh, needleman, blast and noalign. The default is needleman." << "\n"; - cout << "The ksize parameter allows you to specify the kmer size for finding most similar template to candidate. The default is 7." << "\n"; - cout << "The match parameter allows you to specify the bonus for having the same base. The default is 1.0." << "\n"; - cout << "The mistmatch parameter allows you to specify the penalty for having different bases. The default is -1.0." << "\n"; - cout << "The gapopen parameter allows you to specify the penalty for opening a gap in an alignment. The default is -1.0." << "\n"; - cout << "The gapextend parameter allows you to specify the penalty for extending a gap in an alignment. The default is -2.0." << "\n"; - cout << "The align.seqs command should be in the following format: " << "\n"; - cout << "align.seqs(fasta=yourTemplateFile, candidate=yourCandidateFile, align=yourAlignmentMethod, search=yourSearchmethod, ksize=yourKmerSize, match=yourMatchBonus, mismatch=yourMismatchpenalty, gapopen=yourGapopenPenalty, gapextend=yourGapExtendPenalty) " << "\n"; - cout << "Example align.seqs(candidate=candidate.fasta, fasta=core.filtered, align=kmer, search=gotoh, ksize=8, match=2.0, mismatch=3.0, gapopen=-2.0, gapextend=-1.0)" << "\n"; - cout << "Note: No spaces between parameter labels (i.e. fasta), '=' and parameters (i.e.yourFastaFile)." << "\n" << "\n"; + m->mothurOut("The align.seqs command reads a file containing sequences and creates an alignment file and a report file.\n"); + m->mothurOut("The align.seqs command parameters are template, candidate, search, ksize, align, match, mismatch, gapopen and gapextend.\n"); + m->mothurOut("The template and candidate parameters are required. You may enter multiple fasta files by separating their names with dashes. ie. fasta=abrecovery.fasta-amzon.fasta \n"); + m->mothurOut("The search parameter allows you to specify the method to find most similar template. Your options are: suffix, kmer and blast. The default is kmer.\n"); + m->mothurOut("The align parameter allows you to specify the alignment method to use. Your options are: gotoh, needleman, blast and noalign. The default is needleman.\n"); + m->mothurOut("The ksize parameter allows you to specify the kmer size for finding most similar template to candidate. The default is 8.\n"); + m->mothurOut("The match parameter allows you to specify the bonus for having the same base. The default is 1.0.\n"); + m->mothurOut("The mistmatch parameter allows you to specify the penalty for having different bases. The default is -1.0.\n"); + m->mothurOut("The gapopen parameter allows you to specify the penalty for opening a gap in an alignment. The default is -2.0.\n"); + m->mothurOut("The gapextend parameter allows you to specify the penalty for extending a gap in an alignment. The default is -1.0.\n"); + m->mothurOut("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.\n"); + m->mothurOut("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.\n"); + m->mothurOut("If the flip parameter is set to true the reverse complement of the sequence is aligned and the better alignment is reported.\n"); + m->mothurOut("The default for the threshold parameter is 0.50, meaning at least 50% of the bases must remain or the sequence is reported as potentially reversed.\n"); + m->mothurOut("The align.seqs command should be in the following format: \n"); + m->mothurOut("align.seqs(template=yourTemplateFile, candidate=yourCandidateFile, align=yourAlignmentMethod, search=yourSearchmethod, ksize=yourKmerSize, match=yourMatchBonus, mismatch=yourMismatchpenalty, gapopen=yourGapopenPenalty, gapextend=yourGapExtendPenalty) \n"); + m->mothurOut("Example align.seqs(candidate=candidate.fasta, template=core.filtered, align=kmer, search=gotoh, ksize=8, match=2.0, mismatch=3.0, gapopen=-2.0, gapextend=-1.0)\n"); + m->mothurOut("Note: No spaces between parameter labels (i.e. candidate), '=' and parameters (i.e.yourFastaFile).\n\n"); } catch(exception& e) { - cout << "Standard Error: " << e.what() << " has occurred in the AlignCommand class Function help. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + m->errorOut(e, "AlignCommand", "help"); exit(1); } - catch(...) { - cout << "An unknown error has occurred in the AlignCommand class function help. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; - exit(1); - } } @@ -144,193 +197,308 @@ void AlignCommand::help(){ int AlignCommand::execute(){ try { if (abort == true) { return 0; } - - srand( (unsigned)time( NULL ) ); //needed to assign names to temporary files - - if(search == "kmer") { templateDB = new KmerDB(templateFileName, kmerSize); } - else if(search == "suffix") { templateDB = new SuffixDB(templateFileName); } - else if(search == "blast") { templateDB = new BlastDB(templateFileName, gapOpen, gapExtend, match, misMatch); } - else { - cout << search << " is not a valid search option. I will run the command using kmer, ksize=8." << endl; kmerSize = 8; - templateDB = new KmerDB(templateFileName, kmerSize); - } + templateDB = new AlignmentDB(templateFileName, search, kmerSize, gapOpen, gapExtend, match, misMatch); + int longestBase = templateDB->getLongestBase(); - if(align == "gotoh") { alignment = new GotohOverlap(gapOpen, gapExtend, match, misMatch, 3000); } - else if(align == "needleman") { alignment = new NeedlemanOverlap(gapOpen, match, misMatch, 3000); } + if(align == "gotoh") { alignment = new GotohOverlap(gapOpen, gapExtend, match, misMatch, longestBase); } + else if(align == "needleman") { alignment = new NeedlemanOverlap(gapOpen, match, misMatch, longestBase); } else if(align == "blast") { alignment = new BlastAlignment(gapOpen, gapExtend, match, misMatch); } else if(align == "noalign") { alignment = new NoAlign(); } else { - cout << align << " is not a valid alignment option. I will run the command using needleman." << endl; - alignment = new NeedlemanOverlap(gapOpen, match, misMatch, 3000); + m->mothurOut(align + " is not a valid alignment option. I will run the command using needleman."); + m->mothurOutEndLine(); + alignment = new NeedlemanOverlap(gapOpen, match, misMatch, longestBase); } + vector outputNames; - string alignFileName = candidateFileName.substr(0,candidateFileName.find_last_of(".")+1) + "align"; - string reportFileName = candidateFileName.substr(0,candidateFileName.find_last_of(".")+1) + "align.report"; - - int numFastaSeqs = 0; - int start = time(NULL); - -#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) - if(processors == 1){ - ifstream inFASTA; - openInputFile(candidateFileName, inFASTA); - numFastaSeqs=count(istreambuf_iterator(inFASTA),istreambuf_iterator(), '>'); - inFASTA.close(); + for (int s = 0; s < candidateFileNames.size(); s++) { + m->mothurOut("Aligning sequences from " + candidateFileNames[s] + " ..." ); m->mothurOutEndLine(); - lines.push_back(new linePair(0, numFastaSeqs)); + if (outputDir == "") { outputDir += hasPath(candidateFileNames[s]); } + string alignFileName = outputDir + getRootName(getSimpleName(candidateFileNames[s])) + "align"; + string reportFileName = outputDir + getRootName(getSimpleName(candidateFileNames[s])) + "align.report"; + string accnosFileName = outputDir + getRootName(getSimpleName(candidateFileNames[s])) + "flip.accnos"; + bool hasAccnos = true; - driver(lines[0], alignFileName, reportFileName); - } - else{ - vector positions; + int numFastaSeqs = 0; + for (int i = 0; i < lines.size(); i++) { delete lines[i]; } lines.clear(); + int start = time(NULL); - ifstream inFASTA; - openInputFile(candidateFileName, inFASTA); - - while(!inFASTA.eof()){ - char c = inFASTA.get(); - if(c == '>'){ positions.push_back(inFASTA.tellg()); } - while (!inFASTA.eof()) { c = inFASTA.get(); if (c == 10 || c == 13){ break; } } // get rest of line if there's any crap there +#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) + if(processors == 1){ + ifstream inFASTA; + openInputFile(candidateFileNames[s], inFASTA); + numFastaSeqs=count(istreambuf_iterator(inFASTA),istreambuf_iterator(), '>'); + inFASTA.close(); + + lines.push_back(new linePair(0, numFastaSeqs)); + + driver(lines[0], alignFileName, reportFileName, accnosFileName, candidateFileNames[s]); + + //delete accnos file if its blank else report to user + if (isBlank(accnosFileName)) { remove(accnosFileName.c_str()); hasAccnos = false; } + else { + m->mothurOut("Some of you sequences generated alignments that eliminated too many bases, a list is provided in " + accnosFileName + "."); + if (!flip) { + m->mothurOut(" If you set the flip parameter to true mothur will try aligning the reverse compliment as well."); + }else{ m->mothurOut(" If the reverse compliment proved to be better it was reported."); } + m->mothurOutEndLine(); + } } + else{ + vector positions; + processIDS.resize(0); + + ifstream inFASTA; + openInputFile(candidateFileNames[s], 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(); + + numFastaSeqs = positions.size(); + + int numSeqsPerProcessor = numFastaSeqs / processors; + + for (int i = 0; i < processors; i++) { + long int startPos = positions[ i * numSeqsPerProcessor ]; + if(i == processors - 1){ + numSeqsPerProcessor = numFastaSeqs - i * numSeqsPerProcessor; + } + lines.push_back(new linePair(startPos, numSeqsPerProcessor)); + } + + createProcesses(alignFileName, reportFileName, accnosFileName, candidateFileNames[s]); + + rename((alignFileName + toString(processIDS[0]) + ".temp").c_str(), alignFileName.c_str()); + rename((reportFileName + toString(processIDS[0]) + ".temp").c_str(), reportFileName.c_str()); + + //append alignment and report files + for(int i=1;i nonBlankAccnosFiles; + //delete blank accnos files generated with multiple processes + for(int i=0;imothurOut("Some of you sequences generated alignments that eliminated too many bases, a list is provided in " + accnosFileName + "."); + if (!flip) { + m->mothurOut(" If you set the flip parameter to true mothur will try aligning the reverse compliment as well."); + }else{ m->mothurOut(" If the reverse compliment proved to be better it was reported."); } + m->mothurOutEndLine(); + }else{ hasAccnos = false; } + } +#else + ifstream inFASTA; + openInputFile(candidateFileNames[s], inFASTA); + numFastaSeqs=count(istreambuf_iterator(inFASTA),istreambuf_iterator(), '>'); inFASTA.close(); - numFastaSeqs = positions.size(); + lines.push_back(new linePair(0, numFastaSeqs)); - int numSeqsPerProcessor = numFastaSeqs / processors; + driver(lines[0], alignFileName, reportFileName, accnosFileName, candidateFileNames[s]); - for (int i = 0; i < processors; i++) { - int startPos = positions[ i * numSeqsPerProcessor ]; - if(i == processors - 1){ - numSeqsPerProcessor = numFastaSeqs - i * numSeqsPerProcessor; - } - lines.push_back(new linePair(startPos, numSeqsPerProcessor)); + //delete accnos file if its blank else report to user + if (isBlank(accnosFileName)) { remove(accnosFileName.c_str()); hasAccnos = false; } + else { + m->mothurOut("Some of you sequences generated alignments that eliminated too many bases, a list is provided in " + accnosFileName + "."); + if (!flip) { + m->mothurOut(" If you set the flip parameter to true mothur will try aligning the reverse compliment as well."); + }else{ m->mothurOut(" If the reverse compliment proved to be better it was reported."); } + m->mothurOutEndLine(); } - createProcesses(alignFileName, reportFileName); - - //append and remove temp files - map::iterator it = processIDS.begin(); - rename((alignFileName + toString(it->second) + ".temp").c_str(), alignFileName.c_str()); - rename((reportFileName + toString(it->second) + ".temp").c_str(), alignFileName.c_str()); - it++; +#endif - for (; it != processIDS.end(); it++) { - appendAlignFiles((alignFileName + toString(it->second) + ".temp"), alignFileName); - remove((alignFileName + toString(it->second) + ".temp").c_str()); - - appendReportFiles((reportFileName + toString(it->second) + ".temp"), reportFileName); - remove((reportFileName + toString(it->second) + ".temp").c_str()); - } + outputNames.push_back(alignFileName); + outputNames.push_back(reportFileName); + if (hasAccnos) { outputNames.push_back(accnosFileName); } + + m->mothurOut("It took " + toString(time(NULL) - start) + " secs to align " + toString(numFastaSeqs) + " sequences."); + m->mothurOutEndLine(); + m->mothurOutEndLine(); } -#else - ifstream inFASTA; - openInputFile(candidateFileName, inFASTA); - numFastaSeqs=count(istreambuf_iterator(inFASTA),istreambuf_iterator(), '>'); - inFASTA.close(); - lines.push_back(new linePair(0, numFastaSeqs)); - driver(lines[0], alignFileName, reportFileName); -#endif - - cout << "It took " << time(NULL) - start << " secs to align " << numFastaSeqs << " sequences" << endl; - cout << endl; - + m->mothurOutEndLine(); + m->mothurOut("Output File Names: "); m->mothurOutEndLine(); + for (int i = 0; i < outputNames.size(); i++) { m->mothurOut(outputNames[i]); m->mothurOutEndLine(); } + m->mothurOutEndLine(); - return 0; } catch(exception& e) { - cout << "Standard Error: " << e.what() << " has occurred in the AlignCommand class Function execute. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + m->errorOut(e, "AlignCommand", "execute"); exit(1); } - catch(...) { - cout << "An unknown error has occurred in the AlignCommand class function execute. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; - exit(1); - } } //********************************************************************************************************************** -int AlignCommand::driver(linePair* line, string alignFName, string reportFName){ +int AlignCommand::driver(linePair* line, string alignFName, string reportFName, string accnosFName, string filename){ try { ofstream alignmentFile; openOutputFile(alignFName, alignmentFile); + + ofstream accnosFile; + openOutputFile(accnosFName, accnosFile); + NastReport report(reportFName); ifstream inFASTA; - openInputFile(candidateFileName, inFASTA); + openInputFile(filename, inFASTA); + inFASTA.seekg(line->start); for(int i=0;inumSeqs;i++){ + + Sequence* candidateSeq = new Sequence(inFASTA); gobble(inFASTA); + int origNumBases = candidateSeq->getNumBases(); + string originalUnaligned = candidateSeq->getUnaligned(); + int numBasesNeeded = origNumBases * threshold; + + if (candidateSeq->getName() != "") { //incase there is a commented sequence at the end of a file + if (candidateSeq->getUnaligned().length() > alignment->getnRows()) { + alignment->resize(candidateSeq->getUnaligned().length()+1); + } + + Sequence temp = templateDB->findClosestSequence(candidateSeq); + Sequence* templateSeq = &temp; + + float searchScore = templateDB->getSearchScore(); + + Nast* nast = new Nast(alignment, candidateSeq, templateSeq); + Sequence* copy; + + Nast* nast2; + bool needToDeleteCopy = false; //this is needed in case you have you enter the ifs below + //since nast does not make a copy of hte sequence passed, and it is used by the reporter below + //you can't delete the copy sequence til after you report, but you may choose not to create it in the first place + //so this bool tells you if you need to delete it + + //if there is a possibility that this sequence should be reversed + if (candidateSeq->getNumBases() < numBasesNeeded) { + + string wasBetter = ""; + //if the user wants you to try the reverse + if (flip) { + //get reverse compliment + copy = new Sequence(candidateSeq->getName(), originalUnaligned); + copy->reverseComplement(); + + //rerun alignment + Sequence temp2 = templateDB->findClosestSequence(copy); + Sequence* templateSeq2 = &temp2; + + searchScore = templateDB->getSearchScore(); + + nast2 = new Nast(alignment, copy, templateSeq2); - Sequence* candidateSeq = new Sequence(inFASTA); - report.setCandidate(candidateSeq); - - Sequence* templateSeq = templateDB->findClosestSequence(candidateSeq); - report.setTemplate(templateSeq); - report.setSearchParameters(search, templateDB->getSearchScore()); - - Nast nast(alignment, candidateSeq, templateSeq); - report.setAlignmentParameters(align, alignment); - report.setNastParameters(nast); - - alignmentFile << '>' << candidateSeq->getName() << '\n' << candidateSeq->getAligned() << endl; - - report.print(); + //check if any better + if (copy->getNumBases() > candidateSeq->getNumBases()) { + candidateSeq->setAligned(copy->getAligned()); //use reverse compliments alignment since its better + templateSeq = templateSeq2; + delete nast; + nast = nast2; + needToDeleteCopy = true; + }else{ + wasBetter = "\treverse complement did NOT produce a better alignment, please check sequence."; + delete nast2; + delete copy; + } + } + + //create accnos file with names + accnosFile << candidateSeq->getName() << wasBetter << endl; + } + + report.setCandidate(candidateSeq); + report.setTemplate(templateSeq); + report.setSearchParameters(search, searchScore); + report.setAlignmentParameters(align, alignment); + report.setNastParameters(*nast); + + alignmentFile << '>' << candidateSeq->getName() << '\n' << candidateSeq->getAligned() << endl; + + report.print(); + delete nast; + if (needToDeleteCopy) { delete copy; } + } + delete candidateSeq; - delete candidateSeq; + //report progress + if((i+1) % 100 == 0){ m->mothurOut(toString(i+1)); m->mothurOutEndLine(); } } - + //report progress + if((line->numSeqs) % 100 != 0){ m->mothurOut(toString(line->numSeqs)); m->mothurOutEndLine(); } + + alignmentFile.close(); + inFASTA.close(); + accnosFile.close(); + return 1; } catch(exception& e) { - cout << "Standard Error: " << e.what() << " has occurred in the AlignCommand class Function driver. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + m->errorOut(e, "AlignCommand", "driver"); exit(1); } - catch(...) { - cout << "An unknown error has occurred in the AlignCommand class function driver. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; - exit(1); - } } /**************************************************************************************************/ -void AlignCommand::createProcesses(string alignFileName, string reportFileName) { +void AlignCommand::createProcesses(string alignFileName, string reportFileName, string accnosFName, string filename) { try { #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) int process = 0; - processIDS.clear(); + // processIDS.resize(0); //loop through and create all the processes you want while (process != processors) { int pid = fork(); - + if (pid > 0) { - processIDS[lines[process]->numSeqs] = pid; //create map from line number to pid so you can append files in correct order later + 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){ - driver(lines[process], alignFileName + toString(getpid()) + ".temp", reportFileName + toString(getpid()) + ".temp"); + driver(lines[process], alignFileName + toString(getpid()) + ".temp", reportFileName + toString(getpid()) + ".temp", accnosFName + toString(getpid()) + ".temp", filename); exit(0); - }else { cout << "unable to spawn the necessary processes." << endl; exit(0); } + }else { m->mothurOut("unable to spawn the necessary processes."); m->mothurOutEndLine(); exit(0); } } //force parent to wait until all the processes are done - for (map::iterator it = processIDS.begin(); it != processIDS.end(); it++) { - int temp = it->second; + for (int i=0;ierrorOut(e, "AlignCommand", "createProcesses"); exit(1); } - catch(...) { - cout << "An unknown error has occurred in the AlignCommand class function createProcesses. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; - exit(1); - } } /**************************************************************************************************/ @@ -338,70 +506,48 @@ void AlignCommand::createProcesses(string alignFileName, string reportFileName) void AlignCommand::appendAlignFiles(string temp, string filename) { try{ - //open output file in append mode ofstream output; - openOutputFileAppend(filename, output); - - //open temp file for reading ifstream input; + openOutputFileAppend(filename, output); openInputFile(temp, input); - string line; - //read input file and write to output file - while(input.eof() != true) { - getline(input, line); //getline removes the newline char - if (line != "") { - output << line << endl; // Appending back newline char - } - } + while(char c = input.get()){ + if(input.eof()) { break; } + else { output << c; } + } input.close(); output.close(); } catch(exception& e) { - cout << "Standard Error: " << e.what() << " has occurred in the DistanceCommand class Function appendFiles. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + m->errorOut(e, "AlignCommand", "appendAlignFiles"); exit(1); } - catch(...) { - cout << "An unknown error has occurred in the DistanceCommand class function appendFiles. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; - exit(1); - } } - -/**************************************************************************************************/ +//********************************************************************************************************************** void AlignCommand::appendReportFiles(string temp, string filename) { try{ - //open output file in append mode ofstream output; - openOutputFileAppend(filename, output); - - //open temp file for reading ifstream input; + openOutputFileAppend(filename, output); openInputFile(temp, input); - while (!input.eof()) { char c = input.get(); if (c == 10 || c == 13){ break; } } // get header line - string line; - //read input file and write to output file - while(input.eof() != true) { - getline(input, line); //getline removes the newline char - if (line != "") { - output << line << endl; // Appending back newline char - } - } + 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; } + } input.close(); output.close(); } catch(exception& e) { - cout << "Standard Error: " << e.what() << " has occurred in the DistanceCommand class Function appendFiles. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + m->errorOut(e, "AlignCommand", "appendReportFiles"); exit(1); } - catch(...) { - cout << "An unknown error has occurred in the DistanceCommand class function appendFiles. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; - exit(1); - } } //**********************************************************************************************************************