X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=aligncommand.cpp;h=7782d1eac5354261c6727e240a33ae2d89984224;hb=e057ab99eca7d45544c83ca6665a042b934b661b;hp=55f0b821b1392f40b1f6ecd2c9414a6f8a04fff2;hpb=648ec37228eb16075ace911dd5a5773cdfe683da;p=mothur.git diff --git a/aligncommand.cpp b/aligncommand.cpp index 55f0b82..7782d1e 100644 --- a/aligncommand.cpp +++ b/aligncommand.cpp @@ -39,35 +39,78 @@ AlignCommand::AlignCommand(string option){ else { //valid paramters for this command - string AlignArray[] = {"template","candidate","search","ksize","align","match","mismatch","gapopen","gapextend", "processors","flip","threshold"}; + 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") { mothurOut("template is a required parameter for the align.seqs command."); mothurOutEndLine(); abort = true; - } - else if (templateFileName == "not open") { abort = true; } + }else if (templateFileName == "not open") { abort = true; } - candidateFileName = validParameter.validFile(parameters, "candidate", true); - if (candidateFileName == "not found") { - mothurOut("candidate is a required parameter for the align.seqs command."); - mothurOutEndLine(); - abort = true; + candidateFileName = validParameter.validFile(parameters, "candidate", false); + if (candidateFileName == "not found") { mothurOut("candidate is a required parameter for the align.seqs command."); 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) { + mothurOut(candidateFileNames[i] + " will be disregarded."); 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) { mothurOut("no valid files."); mothurOutEndLine(); abort = true; } } - else if (candidateFileName == "not open") { abort = true; } //check for optional parameter and set defaults // ...at some point should added some additional type checking... @@ -93,7 +136,7 @@ AlignCommand::AlignCommand(string option){ 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 = "0.80"; } + 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"; } @@ -125,18 +168,18 @@ void AlignCommand::help(){ try { mothurOut("The align.seqs command reads a file containing sequences and creates an alignment file and a report file.\n"); mothurOut("The align.seqs command parameters are template, candidate, search, ksize, align, match, mismatch, gapopen and gapextend.\n"); - mothurOut("The template and candidate parameters are required.\n"); + 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"); 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"); 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"); mothurOut("The ksize parameter allows you to specify the kmer size for finding most similar template to candidate. The default is 8.\n"); mothurOut("The match parameter allows you to specify the bonus for having the same base. The default is 1.0.\n"); mothurOut("The mistmatch parameter allows you to specify the penalty for having different bases. The default is -1.0.\n"); - mothurOut("The gapopen parameter allows you to specify the penalty for opening a gap in an alignment. The default is -1.0.\n"); - mothurOut("The gapextend parameter allows you to specify the penalty for extending a gap in an alignment. The default is -2.0.\n"); - mothurOut("The flip parameter is used to specify whether or not you want mothur to try the reverse compement if a sequence falls below the threshold. The default is false.\n"); - mothurOut("The threshold is used to specify a cutoff at which an alignment is deemed 'bad' and the reverse complement may be tried. \n"); + mothurOut("The gapopen parameter allows you to specify the penalty for opening a gap in an alignment. The default is -2.0.\n"); + mothurOut("The gapextend parameter allows you to specify the penalty for extending a gap in an alignment. The default is -1.0.\n"); + 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"); + 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"); mothurOut("If the flip parameter is set to true the reverse complement of the sequence is aligned and the better alignment is reported.\n"); - mothurOut("The default for the threshold parameter is 0.80, meaning at least 80% of the bases must remain or the sequence is reported as potentially reversed.\n"); + 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"); mothurOut("The align.seqs command should be in the following format: \n"); mothurOut("align.seqs(template=yourTemplateFile, candidate=yourCandidateFile, align=yourAlignmentMethod, search=yourSearchmethod, ksize=yourKmerSize, match=yourMatchBonus, mismatch=yourMismatchpenalty, gapopen=yourGapopenPenalty, gapextend=yourGapExtendPenalty) \n"); 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"); @@ -167,26 +210,114 @@ int AlignCommand::execute(){ mothurOutEndLine(); alignment = new NeedlemanOverlap(gapOpen, match, misMatch, longestBase); } - mothurOut("Aligning sequences..."); - mothurOutEndLine(); - - string alignFileName = candidateFileName.substr(0,candidateFileName.find_last_of(".")+1) + "align"; - string reportFileName = candidateFileName.substr(0,candidateFileName.find_last_of(".")+1) + "align.report"; - string accnosFileName = candidateFileName.substr(0,candidateFileName.find_last_of(".")+1) + "flip.accnos"; - - int numFastaSeqs = 0; - int start = time(NULL); + for (int s = 0; s < candidateFileNames.size(); s++) { + mothurOut("Aligning sequences from " + candidateFileNames[s] + " ..." ); mothurOutEndLine(); + + 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"; + + int numFastaSeqs = 0; + for (int i = 0; i < lines.size(); i++) { delete lines[i]; } lines.clear(); + int start = time(NULL); + #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) - if(processors == 1){ + 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()); } + else { + mothurOut("Some of you sequences generated alignments that eliminated too many bases, a list is provided in " + accnosFileName + "."); + if (!flip) { + mothurOut(" If you set the flip parameter to true mothur will try aligning the reverse compliment as well."); + }else{ mothurOut(" If the reverse compliment proved to be better it was reported."); } + 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;i(inFASTA),istreambuf_iterator(), '>'); inFASTA.close(); lines.push_back(new linePair(0, numFastaSeqs)); - - driver(lines[0], alignFileName, reportFileName, accnosFileName); + + 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()); } @@ -197,100 +328,16 @@ int AlignCommand::execute(){ }else{ mothurOut(" If the reverse compliment proved to be better it was reported."); } mothurOutEndLine(); } - } - else{ - vector positions; - processIDS.resize(0); - - ifstream inFASTA; - openInputFile(candidateFileName, inFASTA); - - string input; - while(!inFASTA.eof()){ - input = getline(inFASTA); - if (input.length() != 0) { - if(input[0] == '>'){ 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++) { - int startPos = positions[ i * numSeqsPerProcessor ]; - if(i == processors - 1){ - numSeqsPerProcessor = numFastaSeqs - i * numSeqsPerProcessor; - } - lines.push_back(new linePair(startPos, numSeqsPerProcessor)); - } - createProcesses(alignFileName, reportFileName, accnosFileName); +#endif - rename((alignFileName + toString(processIDS[0]) + ".temp").c_str(), alignFileName.c_str()); - rename((reportFileName + toString(processIDS[0]) + ".temp").c_str(), reportFileName.c_str()); - vector nonBlankAccnosFiles; - //delete blank accnos files generated with multiple processes - for(int i=0;i(inFASTA),istreambuf_iterator(), '>'); - inFASTA.close(); - - lines.push_back(new linePair(0, numFastaSeqs)); - - driver(lines[0], alignFileName, reportFileName, accnosFileName); - - //delete accnos file if its blank else report to user - if (isBlank(accnosFileName)) { remove(accnosFileName.c_str()); } - else { - mothurOut("Some of you sequences generated alignments that eliminated too many bases, a list is provided in " + accnosFileName + "."); - if (!flip) { - mothurOut(" If you set the flip parameter to true mothur will try aligning the reverse compliment as well."); - }else{ mothurOut(" If the reverse compliment proved to be better it was reported."); } + mothurOut("It took " + toString(time(NULL) - start) + " secs to align " + toString(numFastaSeqs) + " sequences."); + mothurOutEndLine(); mothurOutEndLine(); } -#endif - - - - mothurOut("It took " + toString(time(NULL) - start) + " secs to align " + toString(numFastaSeqs) + " sequences."); - mothurOutEndLine(); - mothurOutEndLine(); - return 0; } catch(exception& e) { @@ -301,7 +348,7 @@ int AlignCommand::execute(){ //********************************************************************************************************************** -int AlignCommand::driver(linePair* line, string alignFName, string reportFName, string accnosFName){ +int AlignCommand::driver(linePair* line, string alignFName, string reportFName, string accnosFName, string filename){ try { ofstream alignmentFile; openOutputFile(alignFName, alignmentFile); @@ -312,17 +359,17 @@ int AlignCommand::driver(linePair* line, string alignFName, string reportFName, 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); + + 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); @@ -391,7 +438,12 @@ int AlignCommand::driver(linePair* line, string alignFName, string reportFName, if (needToDeleteCopy) { delete copy; } } delete candidateSeq; + + //report progress + if((i+1) % 100 == 0){ mothurOut(toString(i+1)); mothurOutEndLine(); } } + //report progress + if((line->numSeqs) % 100 != 0){ mothurOut(toString(line->numSeqs)); mothurOutEndLine(); } alignmentFile.close(); inFASTA.close(); @@ -407,7 +459,7 @@ int AlignCommand::driver(linePair* line, string alignFName, string reportFName, /**************************************************************************************************/ -void AlignCommand::createProcesses(string alignFileName, string reportFileName, string accnosFName) { +void AlignCommand::createProcesses(string alignFileName, string reportFileName, string accnosFName, string filename) { try { #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) int process = 0; @@ -421,7 +473,7 @@ void AlignCommand::createProcesses(string alignFileName, string reportFileName, 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", accnosFName + toString(getpid()) + ".temp"); + driver(lines[process], alignFileName + toString(getpid()) + ".temp", reportFileName + toString(getpid()) + ".temp", accnosFName + toString(getpid()) + ".temp", filename); exit(0); }else { mothurOut("unable to spawn the necessary processes."); mothurOutEndLine(); exit(0); } }