X-Git-Url: https://git.donarmstrong.com/?p=mothur.git;a=blobdiff_plain;f=classifyseqscommand.cpp;h=1b56433073d404e2c7f33513f7362f8306afcb42;hp=c76b047cbd064a5b7b967ab388109bb4e46cc0cc;hb=cf9987b67aa49777a4c91c2d21f96e58bf17aa82;hpb=49d2b7459c5027557564b21e9487dadafbbbdc96 diff --git a/classifyseqscommand.cpp b/classifyseqscommand.cpp index c76b047..1b56433 100644 --- a/classifyseqscommand.cpp +++ b/classifyseqscommand.cpp @@ -14,27 +14,29 @@ //********************************************************************************************************************** vector ClassifySeqsCommand::setParameters(){ try { - CommandParameter ptaxonomy("taxonomy", "InputTypes", "", "", "none", "none", "none",false,true); parameters.push_back(ptaxonomy); - CommandParameter ptemplate("reference", "InputTypes", "", "", "none", "none", "none",false,true); parameters.push_back(ptemplate); - CommandParameter pfasta("fasta", "InputTypes", "", "", "none", "none", "none",false,true); parameters.push_back(pfasta); - CommandParameter pname("name", "InputTypes", "", "", "none", "none", "none",false,false); parameters.push_back(pname); - CommandParameter pgroup("group", "InputTypes", "", "", "none", "none", "none",false,false); parameters.push_back(pgroup); - CommandParameter psearch("search", "Multiple", "kmer-blast-suffix-distance", "kmer", "", "", "",false,false); parameters.push_back(psearch); - CommandParameter pksize("ksize", "Number", "", "8", "", "", "",false,false); parameters.push_back(pksize); - CommandParameter pmethod("method", "Multiple", "bayesian-knn", "bayesian", "", "", "",false,false); parameters.push_back(pmethod); - CommandParameter pprocessors("processors", "Number", "", "1", "", "", "",false,false); parameters.push_back(pprocessors); - CommandParameter pmatch("match", "Number", "", "1.0", "", "", "",false,false); parameters.push_back(pmatch); - 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 pflip("flip", "Boolean", "", "F", "", "", "",false,false); parameters.push_back(pflip); - CommandParameter pcutoff("cutoff", "Number", "", "0", "", "", "",false,true); parameters.push_back(pcutoff); - CommandParameter pprobs("probs", "Boolean", "", "T", "", "", "",false,false); parameters.push_back(pprobs); - CommandParameter piters("iters", "Number", "", "100", "", "", "",false,true); parameters.push_back(piters); - CommandParameter psave("save", "Boolean", "", "F", "", "", "",false,false); parameters.push_back(psave); - CommandParameter pnumwanted("numwanted", "Number", "", "10", "", "", "",false,true); parameters.push_back(pnumwanted); - CommandParameter pinputdir("inputdir", "String", "", "", "", "", "",false,false); parameters.push_back(pinputdir); - CommandParameter poutputdir("outputdir", "String", "", "", "", "", "",false,false); parameters.push_back(poutputdir); + CommandParameter ptaxonomy("taxonomy", "InputTypes", "", "", "none", "none", "none","",false,true,true); parameters.push_back(ptaxonomy); + CommandParameter ptemplate("reference", "InputTypes", "", "", "none", "none", "none","",false,true,true); parameters.push_back(ptemplate); + CommandParameter pfasta("fasta", "InputTypes", "", "", "none", "none", "none","taxonomy",false,true,true); parameters.push_back(pfasta); + CommandParameter pname("name", "InputTypes", "", "", "NameCount", "none", "none","",false,false,true); parameters.push_back(pname); + CommandParameter pcount("count", "InputTypes", "", "", "NameCount-CountGroup", "none", "none","",false,false,true); parameters.push_back(pcount); + CommandParameter pgroup("group", "InputTypes", "", "", "CountGroup", "none", "none","",false,false,true); parameters.push_back(pgroup); + + CommandParameter psearch("search", "Multiple", "kmer-blast-suffix-distance-align", "kmer", "", "", "","",false,false); parameters.push_back(psearch); + CommandParameter pksize("ksize", "Number", "", "8", "", "", "","",false,false); parameters.push_back(pksize); + CommandParameter pmethod("method", "Multiple", "wang-knn-zap", "wang", "", "", "","",false,false); parameters.push_back(pmethod); + CommandParameter pprocessors("processors", "Number", "", "1", "", "", "","",false,false,true); parameters.push_back(pprocessors); + CommandParameter pmatch("match", "Number", "", "1.0", "", "", "","",false,false); parameters.push_back(pmatch); + 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 pcutoff("cutoff", "Number", "", "0", "", "", "","",false,true); parameters.push_back(pcutoff); + CommandParameter pprobs("probs", "Boolean", "", "T", "", "", "","",false,false); parameters.push_back(pprobs); + CommandParameter piters("iters", "Number", "", "100", "", "", "","",false,true); parameters.push_back(piters); + CommandParameter psave("save", "Boolean", "", "F", "", "", "","",false,false); parameters.push_back(psave); + CommandParameter pshortcuts("shortcuts", "Boolean", "", "T", "", "", "","",false,false); parameters.push_back(pshortcuts); + CommandParameter pnumwanted("numwanted", "Number", "", "10", "", "", "","",false,true); parameters.push_back(pnumwanted); + 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); } @@ -50,12 +52,13 @@ string ClassifySeqsCommand::getHelpString(){ try { string helpString = ""; helpString += "The classify.seqs command reads a fasta file containing sequences and creates a .taxonomy file and a .tax.summary file.\n"; - helpString += "The classify.seqs command parameters are reference, fasta, name, search, ksize, method, taxonomy, processors, match, mismatch, gapopen, gapextend, numwanted and probs.\n"; + helpString += "The classify.seqs command parameters are reference, fasta, name, group, count, search, ksize, method, taxonomy, processors, match, mismatch, gapopen, gapextend, numwanted and probs.\n"; helpString += "The reference, fasta and taxonomy parameters are required. You may enter multiple fasta files by separating their names with dashes. ie. fasta=abrecovery.fasta-amzon.fasta \n"; - helpString += "The search parameter allows you to specify the method to find most similar template. Your options are: suffix, kmer, blast and distance. The default is kmer.\n"; + helpString += "The search parameter allows you to specify the method to find most similar template. Your options are: suffix, kmer, blast, align and distance. The default is kmer.\n"; helpString += "The name parameter allows you add a names file with your fasta file, if you enter multiple fasta files, you must enter matching names files for them.\n"; helpString += "The group parameter allows you add a group file so you can have the summary totals broken up by group.\n"; - helpString += "The method parameter allows you to specify classification method to use. Your options are: bayesian and knn. The default is bayesian.\n"; + helpString += "The count parameter allows you add a count file so you can have the summary totals broken up by group.\n"; + helpString += "The method parameter allows you to specify classification method to use. Your options are: wang, knn and zap. The default is wang.\n"; helpString += "The ksize parameter allows you to specify the kmer size for finding most similar template to candidate. The default is 8.\n"; helpString += "The processors parameter allows you to specify the number of processors to use. The default is 1.\n"; #ifdef USE_MPI @@ -68,8 +71,8 @@ string ClassifySeqsCommand::getHelpString(){ helpString += "The gapextend parameter allows you to specify the penalty for extending a gap in an alignment. The default is -1.0.\n"; helpString += "The numwanted parameter allows you to specify the number of sequence matches you want with the knn method. The default is 10.\n"; helpString += "The cutoff parameter allows you to specify a bootstrap confidence threshold for your taxonomy. The default is 0.\n"; - helpString += "The probs parameter shuts off the bootstrapping results for the bayesian method. The default is true, meaning you want the bootstrapping to be shown.\n"; - helpString += "The iters parameter allows you to specify how many iterations to do when calculating the bootstrap confidence score for your taxonomy with the bayesian method. The default is 100.\n"; + helpString += "The probs parameter shuts off the bootstrapping results for the wang and zap method. The default is true, meaning you want the bootstrapping to be shown.\n"; + helpString += "The iters parameter allows you to specify how many iterations to do when calculating the bootstrap confidence score for your taxonomy with the wang method. The default is 100.\n"; //helpString += "The flip parameter allows you shut off mothur's The default is T.\n"; helpString += "The classify.seqs command should be in the following format: \n"; helpString += "classify.seqs(reference=yourTemplateFile, fasta=yourFastaFile, method=yourClassificationMethod, search=yourSearchmethod, ksize=yourKmerSize, taxonomy=yourTaxonomyFile, processors=yourProcessors) \n"; @@ -85,27 +88,22 @@ string ClassifySeqsCommand::getHelpString(){ } } //********************************************************************************************************************** -string ClassifySeqsCommand::getOutputFileNameTag(string type, string inputName=""){ - try { - string outputFileName = ""; - map >::iterator it; +string ClassifySeqsCommand::getOutputPattern(string type) { + try { + string pattern = ""; - //is this a type this command creates - it = outputTypes.find(type); - if (it == outputTypes.end()) { m->mothurOut("[ERROR]: this command doesn't create a " + type + " output file.\n"); } - else { - if (type == "taxonomy") { outputFileName = "taxonomy"; } - else if (type == "accnos") { outputFileName = "flip.accnos"; } - else if (type == "taxsummary") { outputFileName = "tax.summary"; } - else if (type == "matchdist") { outputFileName = "match.dist"; } - else { m->mothurOut("[ERROR]: No definition for type " + type + " output file tag.\n"); m->control_pressed = true; } - } - return outputFileName; - } - catch(exception& e) { - m->errorOut(e, "ClassifySeqsCommand", "getOutputFileNameTag"); - exit(1); - } + if (type == "taxonomy") { pattern = "[filename],[tag],[tag2],taxonomy"; } + else if (type == "taxsummary") { pattern = "[filename],[tag],[tag2],tax.summary"; } + else if (type == "accnos") { pattern = "[filename],[tag],[tag2],flip.accnos"; } + else if (type == "matchdist") { pattern = "[filename],[tag],[tag2],match.dist"; } + else { m->mothurOut("[ERROR]: No definition for type " + type + " output pattern.\n"); m->control_pressed = true; } + + return pattern; + } + catch(exception& e) { + m->errorOut(e, "ClassifySeqsCommand", "getOutputPattern"); + exit(1); + } } //********************************************************************************************************************** ClassifySeqsCommand::ClassifySeqsCommand(){ @@ -127,7 +125,7 @@ ClassifySeqsCommand::ClassifySeqsCommand(){ ClassifySeqsCommand::ClassifySeqsCommand(string option) { try { abort = false; calledHelp = false; - rdb = ReferenceDB::getInstance(); + rdb = ReferenceDB::getInstance(); hasName = false; hasCount=false; //allow user to run help if(option == "help") { help(); abort = true; calledHelp = true; } @@ -177,15 +175,7 @@ ClassifySeqsCommand::ClassifySeqsCommand(string option) { //if the user has not given a path then, add inputdir. else leave path alone. if (path == "") { parameters["taxonomy"] = inputDir + it->second; } } - - it = parameters.find("group"); - //user has given a template file - if(it != parameters.end()){ - path = m->hasPath(it->second); - //if the user has not given a path then, add inputdir. else leave path alone. - if (path == "") { parameters["group"] = inputDir + it->second; } - } - } + } fastaFileName = validParameter.validFile(parameters, "fasta", false); if (fastaFileName == "not found") { @@ -268,7 +258,6 @@ ClassifySeqsCommand::ClassifySeqsCommand(string option) { namefile = validParameter.validFile(parameters, "name", false); if (namefile == "not found") { namefile = ""; } - else { m->splitAtDash(namefile, namefileNames); @@ -333,11 +322,90 @@ ClassifySeqsCommand::ClassifySeqsCommand(string option) { } } } - + + if (namefileNames.size() != 0) { hasName = true; } + if (namefile != "") { if (namefileNames.size() != fastaFileNames.size()) { abort = true; m->mothurOut("If you provide a name file, you must have one for each fasta file."); m->mothurOutEndLine(); } } + //check for required parameters + countfile = validParameter.validFile(parameters, "count", false); + if (countfile == "not found") { + countfile = ""; + }else { + m->splitAtDash(countfile, countfileNames); + + //go through files and make sure they are good, if not, then disregard them + for (int i = 0; i < countfileNames.size(); i++) { + + bool ignore = false; + if (countfileNames[i] == "current") { + countfileNames[i] = m->getCountTableFile(); + if (countfileNames[i] != "") { m->mothurOut("Using " + countfileNames[i] + " as input file for the count parameter where you had given current."); m->mothurOutEndLine(); } + else { + m->mothurOut("You have no current count file, ignoring current."); m->mothurOutEndLine(); ignore=true; + //erase from file list + countfileNames.erase(countfileNames.begin()+i); + i--; + } + } + + if (!ignore) { + + if (inputDir != "") { + string path = m->hasPath(countfileNames[i]); + //if the user has not given a path then, add inputdir. else leave path alone. + if (path == "") { countfileNames[i] = inputDir + countfileNames[i]; } + } + + int ableToOpen; + ifstream in; + + ableToOpen = m->openInputFile(countfileNames[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(countfileNames[i]); + m->mothurOut("Unable to open " + countfileNames[i] + ". Trying default " + tryPath); m->mothurOutEndLine(); + ifstream in2; + ableToOpen = m->openInputFile(tryPath, in2, "noerror"); + in2.close(); + countfileNames[i] = tryPath; + } + } + + if (ableToOpen == 1) { + if (m->getOutputDir() != "") { //default path is set + string tryPath = m->getOutputDir() + m->getSimpleName(countfileNames[i]); + m->mothurOut("Unable to open " + countfileNames[i] + ". Trying output directory " + tryPath); m->mothurOutEndLine(); + ifstream in2; + ableToOpen = m->openInputFile(tryPath, in2, "noerror"); + in2.close(); + countfileNames[i] = tryPath; + } + } + + in.close(); + + if (ableToOpen == 1) { + m->mothurOut("Unable to open " + countfileNames[i] + ". It will be disregarded."); m->mothurOutEndLine(); + //erase from file list + countfileNames.erase(countfileNames.begin()+i); + i--; + }else { + m->setCountTableFile(countfileNames[i]); + } + } + } + } + + if (countfileNames.size() != 0) { hasCount = true; if (countfileNames.size() != fastaFileNames.size()) {m->mothurOut("If you provide a count file, you must have one for each fasta file."); m->mothurOutEndLine(); } } + + //make sure there is at least one valid file left + if (hasName && hasCount) { m->mothurOut("[ERROR]: You must enter ONLY ONE of the following: count or name."); m->mothurOutEndLine(); abort = true; } + groupfile = validParameter.validFile(parameters, "group", false); if (groupfile == "not found") { groupfile = ""; } else { @@ -345,54 +413,74 @@ ClassifySeqsCommand::ClassifySeqsCommand(string option) { //go through files and make sure they are good, if not, then disregard them for (int i = 0; i < groupfileNames.size(); i++) { - if (inputDir != "") { - string path = m->hasPath(groupfileNames[i]); - //if the user has not given a path then, add inputdir. else leave path alone. - if (path == "") { groupfileNames[i] = inputDir + groupfileNames[i]; } - } - int ableToOpen; - ifstream in; - ableToOpen = m->openInputFile(groupfileNames[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(groupfileNames[i]); - m->mothurOut("Unable to open " + groupfileNames[i] + ". Trying default " + tryPath); m->mothurOutEndLine(); - ifstream in2; - ableToOpen = m->openInputFile(tryPath, in2, "noerror"); - in2.close(); - groupfileNames[i] = tryPath; + bool ignore = false; + if (groupfileNames[i] == "current") { + groupfileNames[i] = m->getGroupFile(); + if (groupfileNames[i] != "") { m->mothurOut("Using " + groupfileNames[i] + " as input file for the group parameter where you had given current."); m->mothurOutEndLine(); } + else { + m->mothurOut("You have no current group file, ignoring current."); m->mothurOutEndLine(); ignore=true; + //erase from file list + groupfileNames.erase(groupfileNames.begin()+i); + i--; } } - if (ableToOpen == 1) { - if (m->getOutputDir() != "") { //default path is set - string tryPath = m->getOutputDir() + m->getSimpleName(groupfileNames[i]); - m->mothurOut("Unable to open " + groupfileNames[i] + ". Trying output directory " + tryPath); m->mothurOutEndLine(); - ifstream in2; - ableToOpen = m->openInputFile(tryPath, in2, "noerror"); - in2.close(); - groupfileNames[i] = tryPath; + if (!ignore) { + + if (inputDir != "") { + string path = m->hasPath(groupfileNames[i]); + cout << path << '\t' << inputDir << endl; + //if the user has not given a path then, add inputdir. else leave path alone. + if (path == "") { groupfileNames[i] = inputDir + groupfileNames[i]; } + } + + int ableToOpen; + + ifstream in; + ableToOpen = m->openInputFile(groupfileNames[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(groupfileNames[i]); + m->mothurOut("Unable to open " + groupfileNames[i] + ". Trying default " + tryPath); m->mothurOutEndLine(); + ifstream in2; + ableToOpen = m->openInputFile(tryPath, in2, "noerror"); + in2.close(); + groupfileNames[i] = tryPath; + } + } + + if (ableToOpen == 1) { + if (m->getOutputDir() != "") { //default path is set + string tryPath = m->getOutputDir() + m->getSimpleName(groupfileNames[i]); + m->mothurOut("Unable to open " + groupfileNames[i] + ". Trying output directory " + tryPath); m->mothurOutEndLine(); + ifstream in2; + ableToOpen = m->openInputFile(tryPath, in2, "noerror"); + in2.close(); + groupfileNames[i] = tryPath; + } + } + + in.close(); + + if (ableToOpen == 1) { + m->mothurOut("Unable to open " + groupfileNames[i] + ". It will be disregarded."); m->mothurOutEndLine(); + //erase from file list + groupfileNames.erase(groupfileNames.begin()+i); + i--; + }else { + m->setGroupFile(groupfileNames[i]); } } - in.close(); - - if (ableToOpen == 1) { - m->mothurOut("Unable to open " + groupfileNames[i] + ". It will be disregarded."); m->mothurOutEndLine(); groupfileNames[i] = ""; - //erase from file list - groupfileNames.erase(groupfileNames.begin()+i); - i--; - }else { - m->setGroupFile(groupfileNames[i]); - } } } if (groupfile != "") { if (groupfileNames.size() != fastaFileNames.size()) { abort = true; m->mothurOut("If you provide a group file, you must have one for each fasta file."); m->mothurOutEndLine(); } + if (hasCount) { m->mothurOut("[ERROR]: You must enter ONLY ONE of the following: count or group."); m->mothurOutEndLine(); abort = true; } }else { for (int i = 0; i < fastaFileNames.size(); i++) { groupfileNames.push_back(""); } } @@ -400,9 +488,6 @@ ClassifySeqsCommand::ClassifySeqsCommand(string option) { //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"; } - m->mothurConvert(temp, kmerSize); - temp = validParameter.validFile(parameters, "processors", false); if (temp == "not found"){ temp = m->getProcessors(); } m->setProcessors(temp); m->mothurConvert(temp, processors); @@ -444,7 +529,13 @@ ClassifySeqsCommand::ClassifySeqsCommand(string option) { search = validParameter.validFile(parameters, "search", false); if (search == "not found"){ search = "kmer"; } - method = validParameter.validFile(parameters, "method", false); if (method == "not found"){ method = "bayesian"; } + method = validParameter.validFile(parameters, "method", false); if (method == "not found"){ method = "wang"; } + + temp = validParameter.validFile(parameters, "ksize", false); if (temp == "not found"){ + temp = "8"; + if (method == "zap") { temp = "7"; } + } + m->mothurConvert(temp, kmerSize); temp = validParameter.validFile(parameters, "match", false); if (temp == "not found"){ temp = "1.0"; } m->mothurConvert(temp, match); @@ -466,6 +557,9 @@ ClassifySeqsCommand::ClassifySeqsCommand(string option) { temp = validParameter.validFile(parameters, "probs", false); if (temp == "not found"){ temp = "true"; } probs = m->isTrue(temp); + + temp = validParameter.validFile(parameters, "shortcuts", false); if (temp == "not found"){ temp = "true"; } + writeShortcuts = m->isTrue(temp); //temp = validParameter.validFile(parameters, "flip", false); if (temp == "not found"){ temp = "T"; } //flip = m->isTrue(temp); @@ -475,16 +569,23 @@ ClassifySeqsCommand::ClassifySeqsCommand(string option) { m->mothurConvert(temp, iters); - if ((method == "bayesian") && (search != "kmer")) { - m->mothurOut("The bayesian method requires the kmer search." + search + "will be disregarded." ); m->mothurOutEndLine(); + if ((method == "wang") && (search != "kmer")) { + m->mothurOut("The wang method requires the kmer search. " + search + " will be disregarded, and kmer will be used." ); m->mothurOutEndLine(); + search = "kmer"; + } + + if ((method == "zap") && ((search != "kmer") && (search != "align"))) { + m->mothurOut("The zap method requires the kmer or align search. " + search + " will be disregarded, and kmer will be used." ); m->mothurOutEndLine(); search = "kmer"; } if (!abort) { - if (namefileNames.size() == 0){ - if (fastaFileNames.size() != 0) { - vector files; files.push_back(fastaFileNames[fastaFileNames.size()-1]); - parser.getNameFile(files); + if (!hasCount) { + if (namefileNames.size() == 0){ + if (fastaFileNames.size() != 0) { + vector files; files.push_back(fastaFileNames[fastaFileNames.size()-1]); + parser.getNameFile(files); + } } } } @@ -508,12 +609,18 @@ int ClassifySeqsCommand::execute(){ try { if (abort == true) { if (calledHelp) { return 0; } return 2; } - if(method == "bayesian"){ classify = new Bayesian(taxonomyFileName, templateFileName, search, kmerSize, cutoff, iters, rand(), flip); } + string outputMethodTag = method; + if(method == "wang"){ classify = new Bayesian(taxonomyFileName, templateFileName, search, kmerSize, cutoff, iters, rand(), flip, writeShortcuts); } else if(method == "knn"){ classify = new Knn(taxonomyFileName, templateFileName, search, kmerSize, gapOpen, gapExtend, match, misMatch, numWanted, rand()); } + else if(method == "zap"){ + outputMethodTag = search + "_" + outputMethodTag; + if (search == "kmer") { classify = new KmerTree(templateFileName, taxonomyFileName, kmerSize, cutoff); } + else { classify = new AlignTree(templateFileName, taxonomyFileName, cutoff); } + } else { - m->mothurOut(search + " is not a valid method option. I will run the command using bayesian."); + m->mothurOut(search + " is not a valid method option. I will run the command using wang."); m->mothurOutEndLine(); - classify = new Bayesian(taxonomyFileName, templateFileName, search, kmerSize, cutoff, iters, rand(), flip); + classify = new Bayesian(taxonomyFileName, templateFileName, search, kmerSize, cutoff, iters, rand(), flip, writeShortcuts); } if (m->control_pressed) { delete classify; return 0; } @@ -522,8 +629,8 @@ int ClassifySeqsCommand::execute(){ m->mothurOut("Classifying sequences from " + fastaFileNames[s] + " ..." ); m->mothurOutEndLine(); - string baseTName = taxonomyFileName; - if (taxonomyFileName == "saved") {baseTName = rdb->getSavedTaxonomy(); } + string baseTName = m->getSimpleName(taxonomyFileName); + if (taxonomyFileName == "saved") { baseTName = rdb->getSavedTaxonomy(); } //set rippedTaxName to string RippedTaxName = ""; @@ -533,21 +640,24 @@ int ClassifySeqsCommand::execute(){ else if (foundDot && (baseTName[i] == '.')) { break; } else if (!foundDot && (baseTName[i] == '.')) { foundDot = true; } } - if (RippedTaxName != "") { RippedTaxName += "."; } + //if (RippedTaxName != "") { RippedTaxName += "."; } if (outputDir == "") { outputDir += m->hasPath(fastaFileNames[s]); } - string newTaxonomyFile = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + RippedTaxName + getOutputFileNameTag("taxonomy"); - string newaccnosFile = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + RippedTaxName + getOutputFileNameTag("accnos"); + map variables; + variables["[filename]"] = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])); + variables["[tag]"] = RippedTaxName; + variables["[tag2]"] = outputMethodTag; + string newTaxonomyFile = getOutputFileName("taxonomy", variables); + string newaccnosFile = getOutputFileName("accnos", variables); string tempTaxonomyFile = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + "taxonomy.temp"; - string taxSummary = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + RippedTaxName + getOutputFileNameTag("taxsummary"); + string taxSummary = getOutputFileName("taxsummary", variables); if ((method == "knn") && (search == "distance")) { - string DistName = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + getOutputFileNameTag("matchdist"); + string DistName = getOutputFileName("matchdist", variables); classify->setDistName(DistName); outputNames.push_back(DistName); outputTypes["matchdist"].push_back(DistName); } outputNames.push_back(newTaxonomyFile); outputTypes["taxonomy"].push_back(newTaxonomyFile); - outputNames.push_back(newaccnosFile); outputTypes["accnos"].push_back(newaccnosFile); outputNames.push_back(taxSummary); outputTypes["taxsummary"].push_back(taxSummary); int start = time(NULL); @@ -671,7 +781,9 @@ int ClassifySeqsCommand::execute(){ } #endif - if (!m->isBlank(newaccnosFile)) { m->mothurOutEndLine(); m->mothurOut("[WARNING]: mothur suspects some of your sequences may be reversed, please check " + newaccnosFile + " for the list of the sequences."); m->mothurOutEndLine(); } + if (!m->isBlank(newaccnosFile)) { m->mothurOutEndLine(); m->mothurOut("[WARNING]: mothur reversed some your sequences for a better classification. If you would like to take a closer look, please check " + newaccnosFile + " for the list of the sequences."); m->mothurOutEndLine(); + outputNames.push_back(newaccnosFile); outputTypes["accnos"].push_back(newaccnosFile); + }else { m->mothurRemove(newaccnosFile); } m->mothurOutEndLine(); m->mothurOut("It took " + toString(time(NULL) - start) + " secs to classify " + toString(numFastaSeqs) + " sequences."); m->mothurOutEndLine(); m->mothurOutEndLine(); @@ -694,46 +806,58 @@ int ClassifySeqsCommand::execute(){ } #endif - string group = ""; - if (groupfile != "") { group = groupfileNames[s]; } - - PhyloSummary taxaSum(baseTName, group); - - if (m->control_pressed) { outputTypes.clear(); for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } delete classify; return 0; } - - if (namefile == "") { taxaSum.summarize(tempTaxonomyFile); } - else { - ifstream in; - m->openInputFile(tempTaxonomyFile, in); - - //read in users taxonomy file and add sequences to tree - string name, taxon; - - while(!in.eof()){ - in >> name >> taxon; m->gobble(in); - - itNames = nameMap.find(name); - - if (itNames == nameMap.end()) { - m->mothurOut(name + " is not in your name file please correct."); m->mothurOutEndLine(); exit(1); - }else{ - for (int i = 0; i < itNames->second.size(); i++) { - taxaSum.addSeqToTree(itNames->second[i], taxon); //add it as many times as there are identical seqs - } - itNames->second.clear(); - nameMap.erase(itNames->first); - } - } - in.close(); - } + string group = ""; + GroupMap* groupMap = NULL; + CountTable* ct = NULL; + PhyloSummary* taxaSum; + if (hasCount) { + ct = new CountTable(); + ct->readTable(countfileNames[s], true, false); + taxaSum = new PhyloSummary(taxonomyFileName, ct); + taxaSum->summarize(tempTaxonomyFile); + }else { + if (groupfile != "") { group = groupfileNames[s]; groupMap = new GroupMap(group); groupMap->readMap(); } + + taxaSum = new PhyloSummary(taxonomyFileName, groupMap); + + if (m->control_pressed) { outputTypes.clear(); if (ct != NULL) { delete ct; } if (groupMap != NULL) { delete groupMap; } delete taxaSum; for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } delete classify; return 0; } + + if (namefile == "") { taxaSum->summarize(tempTaxonomyFile); } + else { + ifstream in; + m->openInputFile(tempTaxonomyFile, in); + + //read in users taxonomy file and add sequences to tree + string name, taxon; + + while(!in.eof()){ + if (m->control_pressed) { outputTypes.clear(); if (ct != NULL) { delete ct; } if (groupMap != NULL) { delete groupMap; } delete taxaSum; for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } delete classify; return 0; } + + in >> name >> taxon; m->gobble(in); + + itNames = nameMap.find(name); + + if (itNames == nameMap.end()) { + m->mothurOut(name + " is not in your name file please correct."); m->mothurOutEndLine(); exit(1); + }else{ + for (int i = 0; i < itNames->second.size(); i++) { + taxaSum->addSeqToTree(itNames->second[i], taxon); //add it as many times as there are identical seqs + } + itNames->second.clear(); + nameMap.erase(itNames->first); + } + } + in.close(); + } + } m->mothurRemove(tempTaxonomyFile); - if (m->control_pressed) { outputTypes.clear(); for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } delete classify; return 0; } + if (m->control_pressed) { outputTypes.clear(); if (ct != NULL) { delete ct; } if (groupMap != NULL) { delete groupMap; } for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } delete classify; return 0; } //print summary file ofstream outTaxTree; m->openOutputFile(taxSummary, outTaxTree); - taxaSum.print(outTaxTree); + taxaSum->print(outTaxTree); outTaxTree.close(); //output taxonomy with the unclassified bins added @@ -745,12 +869,12 @@ int ClassifySeqsCommand::execute(){ m->openOutputFile(unclass, outTax); //get maxLevel from phylotree so you know how many 'unclassified's to add - int maxLevel = taxaSum.getMaxLevel(); + int maxLevel = taxaSum->getMaxLevel(); //read taxfile - this reading and rewriting is done to preserve the confidence scores. string name, taxon; while (!inTax.eof()) { - if (m->control_pressed) { outputTypes.clear(); for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } m->mothurRemove(unclass); delete classify; return 0; } + if (m->control_pressed) { outputTypes.clear(); if (ct != NULL) { delete ct; } if (groupMap != NULL) { delete groupMap; } delete taxaSum; for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } m->mothurRemove(unclass); delete classify; return 0; } inTax >> name >> taxon; m->gobble(inTax); @@ -761,6 +885,8 @@ int ClassifySeqsCommand::execute(){ inTax.close(); outTax.close(); + if (ct != NULL) { delete ct; } + if (groupMap != NULL) { delete groupMap; } delete taxaSum; m->mothurRemove(newTaxonomyFile); rename(unclass.c_str(), newTaxonomyFile.c_str()); @@ -770,12 +896,13 @@ int ClassifySeqsCommand::execute(){ #ifdef USE_MPI } #endif - - 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(); } + delete classify; + + 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(); //set taxonomy file as new current taxonomyfile string current = ""; @@ -790,7 +917,7 @@ int ClassifySeqsCommand::execute(){ if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setAccnosFile(current); } } - delete classify; + return 0; } @@ -897,7 +1024,7 @@ int ClassifySeqsCommand::createProcesses(string taxFileName, string tempTaxFile, string extension = ""; if (i != 0) { extension = toString(i) + ".temp"; processIDS.push_back(i); } - classifyData* tempclass = new classifyData((accnos + extension), probs, method, templateFileName, taxonomyFileName, (taxFileName + extension), (tempTaxFile + extension), filename, search, kmerSize, iters, numWanted, m, lines[i]->start, lines[i]->end, match, misMatch, gapOpen, gapExtend, cutoff, i, flip); + classifyData* tempclass = new classifyData((accnos + extension), probs, method, templateFileName, taxonomyFileName, (taxFileName + extension), (tempTaxFile + extension), filename, search, kmerSize, iters, numWanted, m, lines[i]->start, lines[i]->end, match, misMatch, gapOpen, gapExtend, cutoff, i, flip, writeShortcuts); pDataArray.push_back(tempclass); //MySeqSumThreadFunction is in header. It must be global or static to work with the threads. @@ -916,6 +1043,9 @@ int ClassifySeqsCommand::createProcesses(string taxFileName, string tempTaxFile, //Close all thread handles and free memory allocations. for(int i=0; i < pDataArray.size(); i++){ num += pDataArray[i]->count; + 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; + } CloseHandle(hThreadArray[i]); delete pDataArray[i]; } @@ -1021,11 +1151,11 @@ int ClassifySeqsCommand::driver(linePair* filePos, string taxFName, string tempT #endif //report progress - if((count) % 100 == 0){ m->mothurOut("Processing sequence: " + toString(count)); m->mothurOutEndLine(); } + if((count) % 100 == 0){ m->mothurOutJustToScreen("Processing sequence: " + toString(count) +"\n"); } } //report progress - if((count) % 100 != 0){ m->mothurOut("Processing sequence: " + toString(count)); m->mothurOutEndLine(); } + if((count) % 100 != 0){ m->mothurOutJustToScreen("Processing sequence: " + toString(count)+"\n"); } inFASTA.close(); outTax.close();