]> git.donarmstrong.com Git - mothur.git/blobdiff - classifyseqscommand.cpp
fixed io
[mothur.git] / classifyseqscommand.cpp
index 9ee7b5e946a7f5819cbac94bab7efe8c0cb71aef..c76b047cbd064a5b7b967ab388109bb4e46cc0cc 100644 (file)
@@ -27,6 +27,7 @@ vector<string> ClassifySeqsCommand::setParameters(){
                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);
@@ -69,6 +70,7 @@ string ClassifySeqsCommand::getHelpString(){
                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 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";
                helpString += "Example classify.seqs(fasta=amazon.fasta, reference=core.filtered, method=knn, search=gotoh, ksize=8, processors=2)\n";
@@ -83,12 +85,36 @@ string ClassifySeqsCommand::getHelpString(){
        }
 }
 //**********************************************************************************************************************
+string ClassifySeqsCommand::getOutputFileNameTag(string type, string inputName=""){    
+       try {
+        string outputFileName = "";
+               map<string, vector<string> >::iterator it;
+        
+        //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);
+       }
+}
+//**********************************************************************************************************************
 ClassifySeqsCommand::ClassifySeqsCommand(){    
        try {
                abort = true; calledHelp = true; 
                setParameters();
                vector<string> tempOutNames;
                outputTypes["taxonomy"] = tempOutNames;
+               outputTypes["accnos"] = tempOutNames;
                outputTypes["taxsummary"] = tempOutNames;
                outputTypes["matchdist"] = tempOutNames;
        }
@@ -126,6 +152,7 @@ ClassifySeqsCommand::ClassifySeqsCommand(string option)  {
                        outputTypes["taxonomy"] = tempOutNames;
                        outputTypes["taxsummary"] = tempOutNames;
                        outputTypes["matchdist"] = tempOutNames;
+                       outputTypes["accnos"] = tempOutNames;
                        
                        //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 = "";         }
@@ -440,17 +467,28 @@ ClassifySeqsCommand::ClassifySeqsCommand(string option)  {
                        temp = validParameter.validFile(parameters, "probs", false);            if (temp == "not found"){       temp = "true";                  }
                        probs = m->isTrue(temp);
                        
+                       //temp = validParameter.validFile(parameters, "flip", false);                   if (temp == "not found"){       temp = "T";                             }
+                       //flip = m->isTrue(temp); 
+                       flip = true;
+                       
                        temp = validParameter.validFile(parameters, "iters", false);            if (temp == "not found") { temp = "100";                        }
                        m->mothurConvert(temp, iters); 
 
-
                        
                        if ((method == "bayesian") && (search != "kmer"))  { 
                                m->mothurOut("The bayesian method requires the kmer search." + search + "will be disregarded." ); m->mothurOutEndLine();
                                search = "kmer";
                        }
-               }
-               
+                       
+            if (!abort) {
+                if (namefileNames.size() == 0){
+                    if (fastaFileNames.size() != 0) {
+                        vector<string> files; files.push_back(fastaFileNames[fastaFileNames.size()-1]); 
+                        parser.getNameFile(files);
+                    }
+                }
+            }
+        }
        }
        catch(exception& e) {
                m->errorOut(e, "ClassifySeqsCommand", "ClassifySeqsCommand");
@@ -469,13 +507,13 @@ ClassifySeqsCommand::~ClassifySeqsCommand(){
 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());           }
+        
+               if(method == "bayesian"){       classify = new Bayesian(taxonomyFileName, templateFileName, search, kmerSize, cutoff, iters, rand(), flip);             }
                else if(method == "knn"){       classify = new Knn(taxonomyFileName, templateFileName, search, kmerSize, gapOpen, gapExtend, match, misMatch, numWanted, rand());                               }
                else {
                        m->mothurOut(search + " is not a valid method option. I will run the command using bayesian.");
                        m->mothurOutEndLine();
-                       classify = new Bayesian(taxonomyFileName, templateFileName, search, kmerSize, cutoff, iters, rand());   
+                       classify = new Bayesian(taxonomyFileName, templateFileName, search, kmerSize, cutoff, iters, rand(), flip);     
                }
                
                if (m->control_pressed) { delete classify; return 0; }
@@ -487,22 +525,29 @@ int ClassifySeqsCommand::execute(){
                        string baseTName = taxonomyFileName;
                        if (taxonomyFileName == "saved") {baseTName = rdb->getSavedTaxonomy();  }
                        
-                       string RippedTaxName = m->getRootName(m->getSimpleName(baseTName));
-                       RippedTaxName = m->getExtension(RippedTaxName.substr(0, RippedTaxName.length()-1));
-                       if (RippedTaxName[0] == '.') { RippedTaxName = RippedTaxName.substr(1, RippedTaxName.length()); }
-                       RippedTaxName +=  "."; 
-               
+            //set rippedTaxName to 
+                       string RippedTaxName = "";
+            bool foundDot = false;
+            for (int i = baseTName.length()-1; i >= 0; i--) {
+                if (foundDot && (baseTName[i] != '.')) {  RippedTaxName = baseTName[i] + RippedTaxName; }
+                else if (foundDot && (baseTName[i] == '.')) {  break; }
+                else if (!foundDot && (baseTName[i] == '.')) {  foundDot = true; }
+            }
+            if (RippedTaxName != "") { RippedTaxName +=  "."; }   
+          
                        if (outputDir == "") { outputDir += m->hasPath(fastaFileNames[s]); }
-                       string newTaxonomyFile = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + RippedTaxName + "taxonomy";
+                       string newTaxonomyFile = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + RippedTaxName + getOutputFileNameTag("taxonomy");
+                       string newaccnosFile = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + RippedTaxName + getOutputFileNameTag("accnos");
                        string tempTaxonomyFile = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + "taxonomy.temp";
-                       string taxSummary = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + RippedTaxName + "tax.summary";
+                       string taxSummary = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + RippedTaxName + getOutputFileNameTag("taxsummary");
                        
                        if ((method == "knn") && (search == "distance")) { 
-                               string DistName = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + "match.dist";
+                               string DistName = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + getOutputFileNameTag("matchdist");
                                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);
@@ -521,6 +566,7 @@ int ClassifySeqsCommand::execute(){
                                MPI_File inMPI;
                                MPI_File outMPINewTax;
                                MPI_File outMPITempTax;
+                               MPI_File outMPIAcc;
                                                        
                                int outMode=MPI_MODE_CREATE|MPI_MODE_WRONLY; 
                                int inMode=MPI_MODE_RDONLY; 
@@ -530,6 +576,9 @@ int ClassifySeqsCommand::execute(){
                                
                                char outTempTax[1024];
                                strcpy(outTempTax, tempTaxonomyFile.c_str());
+                       
+                               char outAcc[1024];
+                               strcpy(outAcc, newaccnosFile.c_str());
                                
                                char inFileName[1024];
                                strcpy(inFileName, fastaFileNames[s].c_str());
@@ -537,8 +586,9 @@ int ClassifySeqsCommand::execute(){
                                MPI_File_open(MPI_COMM_WORLD, inFileName, inMode, MPI_INFO_NULL, &inMPI);  //comm, filename, mode, info, filepointer
                                MPI_File_open(MPI_COMM_WORLD, outNewTax, outMode, MPI_INFO_NULL, &outMPINewTax);
                                MPI_File_open(MPI_COMM_WORLD, outTempTax, outMode, MPI_INFO_NULL, &outMPITempTax);
+                               MPI_File_open(MPI_COMM_WORLD, outAcc, outMode, MPI_INFO_NULL, &outMPIAcc);
                                
-                               if (m->control_pressed) { outputTypes.clear(); MPI_File_close(&inMPI);  MPI_File_close(&outMPINewTax);   MPI_File_close(&outMPITempTax);  delete classify;  return 0;  }
+                               if (m->control_pressed) { outputTypes.clear(); MPI_File_close(&inMPI);  MPI_File_close(&outMPINewTax);  MPI_File_close(&outMPIAcc);   MPI_File_close(&outMPITempTax);  delete classify;  return 0;  }
                                
                                if (pid == 0) { //you are the root process 
                                        
@@ -557,9 +607,9 @@ int ClassifySeqsCommand::execute(){
                                        
                                
                                        //align your part
-                                       driverMPI(startIndex, numSeqsPerProcessor, inMPI, outMPINewTax, outMPITempTax, MPIPos);
+                                       driverMPI(startIndex, numSeqsPerProcessor, inMPI, outMPINewTax, outMPITempTax, outMPIAcc, MPIPos);
                                        
-                                       if (m->control_pressed) {  outputTypes.clear(); MPI_File_close(&inMPI);  MPI_File_close(&outMPINewTax);   MPI_File_close(&outMPITempTax);  for (int i = 0; i < outputNames.size(); i++) {       m->mothurRemove(outputNames[i]);        } delete classify; return 0;  }
+                                       if (m->control_pressed) {  outputTypes.clear(); MPI_File_close(&inMPI);  MPI_File_close(&outMPINewTax); MPI_File_close(&outMPIAcc);   MPI_File_close(&outMPITempTax);  for (int i = 0; i < outputNames.size(); i++) {   m->mothurRemove(outputNames[i]);        } delete classify; return 0;  }
                                        
                                        for (int i = 1; i < processors; i++) {
                                                int done;
@@ -577,9 +627,9 @@ int ClassifySeqsCommand::execute(){
                                        
                                        
                                        //align your part
-                                       driverMPI(startIndex, numSeqsPerProcessor, inMPI, outMPINewTax, outMPITempTax, MPIPos);
+                                       driverMPI(startIndex, numSeqsPerProcessor, inMPI, outMPINewTax, outMPITempTax, outMPIAcc, MPIPos);
                                        
-                                       if (m->control_pressed) {  outputTypes.clear(); MPI_File_close(&inMPI);  MPI_File_close(&outMPINewTax);   MPI_File_close(&outMPITempTax);  delete classify; return 0;  }
+                                       if (m->control_pressed) {  outputTypes.clear(); MPI_File_close(&inMPI);  MPI_File_close(&outMPINewTax);  MPI_File_close(&outMPIAcc);  MPI_File_close(&outMPITempTax);  delete classify; return 0;  }
 
                                        int done = 0;
                                        MPI_Send(&done, 1, MPI_INT, 0, tag, MPI_COMM_WORLD); 
@@ -589,12 +639,13 @@ int ClassifySeqsCommand::execute(){
                                MPI_File_close(&inMPI);
                                MPI_File_close(&outMPINewTax);
                                MPI_File_close(&outMPITempTax);
+                               MPI_File_close(&outMPIAcc); 
                                MPI_Barrier(MPI_COMM_WORLD); //make everyone wait - just in case
                                
 #else
                
                        vector<unsigned long long> positions; 
-#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
+#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) || (__linux__) || (__unix__) || (__unix)
                        positions = m->divideFile(fastaFileNames[s], processors);
                        for (int i = 0; i < (positions.size()-1); i++) {        lines.push_back(new linePair(positions[i], positions[(i+1)]));  }
 #else
@@ -602,6 +653,7 @@ int ClassifySeqsCommand::execute(){
                                lines.push_back(new linePair(0, 1000));
                        }else {
                                positions = m->setFilePosFasta(fastaFileNames[s], numFastaSeqs); 
+                if (positions.size() < processors) { processors = positions.size(); }
                                
                                //figure out how many sequences you have to process
                                int numSeqsPerProcessor = numFastaSeqs / processors;
@@ -613,16 +665,19 @@ int ClassifySeqsCommand::execute(){
                        }
 #endif
                        if(processors == 1){
-                               numFastaSeqs = driver(lines[0], newTaxonomyFile, tempTaxonomyFile, fastaFileNames[s]);
+                               numFastaSeqs = driver(lines[0], newTaxonomyFile, tempTaxonomyFile, newaccnosFile, fastaFileNames[s]);
                        }else{
-                               numFastaSeqs = createProcesses(newTaxonomyFile, tempTaxonomyFile, fastaFileNames[s]); 
+                               numFastaSeqs = createProcesses(newTaxonomyFile, tempTaxonomyFile, newaccnosFile, fastaFileNames[s]); 
                        }
 #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(); }
 
                m->mothurOutEndLine();
                m->mothurOut("It took " + toString(time(NULL) - start) + " secs to classify " + toString(numFastaSeqs) + " sequences."); m->mothurOutEndLine(); m->mothurOutEndLine();
                start = time(NULL);
-
+               
+               
 
                #ifdef USE_MPI  
                        if (pid == 0) {  //this part does not need to be paralellized
@@ -633,23 +688,8 @@ int ClassifySeqsCommand::execute(){
                        if(namefile != "") {
                        
                            m->mothurOut("Reading " + namefileNames[s] + "..."); cout.flush();
-                               
                                nameMap.clear(); //remove old names
-                               
-                               ifstream inNames;
-                               m->openInputFile(namefileNames[s], inNames);
-                               
-                               string firstCol, secondCol;
-                               while(!inNames.eof()) {
-                                       inNames >> firstCol >> secondCol; m->gobble(inNames);
-                                       
-                                       vector<string> temp;
-                                       m->splitAtComma(secondCol, temp);
-                       
-                                       nameMap[firstCol] = temp;  
-                               }
-                               inNames.close();
-                               
+                               m->readNames(namefileNames[s], nameMap);                                
                                m->mothurOut("  Done."); m->mothurOutEndLine();
                        }
                #endif
@@ -744,6 +784,12 @@ int ClassifySeqsCommand::execute(){
                        if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setTaxonomyFile(current); }
                }
                
+               current = "";
+               itTypes = outputTypes.find("accnos");
+               if (itTypes != outputTypes.end()) {
+                       if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setAccnosFile(current); }
+               }
+               
                delete classify;
                
                return 0;
@@ -785,13 +831,13 @@ string ClassifySeqsCommand::addUnclassifieds(string tax, int maxlevel) {
 
 /**************************************************************************************************/
 
-int ClassifySeqsCommand::createProcesses(string taxFileName, string tempTaxFile, string filename) {
+int ClassifySeqsCommand::createProcesses(string taxFileName, string tempTaxFile, string accnos, string filename) {
        try {
                
                int num = 0;
                processIDS.clear();
                
-#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
@@ -802,7 +848,7 @@ int ClassifySeqsCommand::createProcesses(string taxFileName, string tempTaxFile,
                                processIDS.push_back(pid);  //create map from line number to pid so you can append files in correct order later
                                process++;
                        }else if (pid == 0){
-                               num = driver(lines[process], taxFileName + toString(getpid()) + ".temp", tempTaxFile + toString(getpid()) + ".temp", filename);
+                               num = driver(lines[process], taxFileName + toString(getpid()) + ".temp", tempTaxFile + toString(getpid()) + ".temp", accnos + toString(getpid()) + ".temp", filename);
 
                                //pass numSeqs to parent
                                ofstream out;
@@ -820,7 +866,7 @@ int ClassifySeqsCommand::createProcesses(string taxFileName, string tempTaxFile,
                }
                
                //parent does its part
-               num = driver(lines[0], taxFileName, tempTaxFile, filename);
+               num = driver(lines[0], taxFileName, tempTaxFile, accnos, filename);
                
                //force parent to wait until all the processes are done
                for (int i=0;i<processIDS.size();i++) { 
@@ -851,7 +897,7 @@ int ClassifySeqsCommand::createProcesses(string taxFileName, string tempTaxFile,
                        string extension = "";
                        if (i != 0) { extension = toString(i) + ".temp"; processIDS.push_back(i); }
                        
-                       classifyData* tempclass = new classifyData(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);
+                       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);
                        pDataArray.push_back(tempclass);
                        
                        //MySeqSumThreadFunction is in header. It must be global or static to work with the threads.
@@ -861,7 +907,7 @@ int ClassifySeqsCommand::createProcesses(string taxFileName, string tempTaxFile,
                }
                
                //parent does its part
-               num = driver(lines[processors-1], taxFileName + toString(processors-1) + ".temp", tempTaxFile + toString(processors-1) + ".temp", filename);
+               num = driver(lines[processors-1], taxFileName + toString(processors-1) + ".temp", tempTaxFile + toString(processors-1) + ".temp", accnos + toString(processors-1) + ".temp", filename);
                processIDS.push_back((processors-1));
                
                //Wait until all threads have terminated.
@@ -875,14 +921,35 @@ int ClassifySeqsCommand::createProcesses(string taxFileName, string tempTaxFile,
                }
                
        #endif  
-               
+        vector<string> nonBlankAccnosFiles;
+               if (!(m->isBlank(accnos))) { nonBlankAccnosFiles.push_back(accnos); }
+               else { m->mothurRemove(accnos); } //remove so other files can be renamed to it
+        
                for(int i=0;i<processIDS.size();i++){
-                       appendTaxFiles((taxFileName + toString(processIDS[i]) + ".temp"), taxFileName);
-                       appendTaxFiles((tempTaxFile + toString(processIDS[i]) + ".temp"), tempTaxFile);
+                       m->appendFiles((taxFileName + toString(processIDS[i]) + ".temp"), taxFileName);
+                       m->appendFiles((tempTaxFile + toString(processIDS[i]) + ".temp"), tempTaxFile);
+            if (!(m->isBlank(accnos + toString(processIDS[i]) + ".temp"))) {
+                               nonBlankAccnosFiles.push_back(accnos + toString(processIDS[i]) + ".temp");
+                       }else { m->mothurRemove((accnos + toString(processIDS[i]) + ".temp"));  }
+
                        m->mothurRemove((m->getFullPathName(taxFileName) + toString(processIDS[i]) + ".temp"));
                        m->mothurRemove((m->getFullPathName(tempTaxFile) + toString(processIDS[i]) + ".temp"));
                }
                
+        //append accnos files
+               if (nonBlankAccnosFiles.size() != 0) { 
+                       rename(nonBlankAccnosFiles[0].c_str(), accnos.c_str());
+                       
+                       for (int h=1; h < nonBlankAccnosFiles.size(); h++) {
+                               m->appendFiles(nonBlankAccnosFiles[h], accnos);
+                               m->mothurRemove(nonBlankAccnosFiles[h]);
+                       }
+               }else { //recreate the accnosfile if needed
+                       ofstream out;
+                       m->openOutputFile(accnos, out);
+                       out.close();
+               }
+
                return num;
                
        }
@@ -891,39 +958,18 @@ int ClassifySeqsCommand::createProcesses(string taxFileName, string tempTaxFile,
                exit(1);
        }
 }
-/**************************************************************************************************/
-
-void ClassifySeqsCommand::appendTaxFiles(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, "ClassifySeqsCommand", "appendTaxFiles");
-               exit(1);
-       }
-}
-
 //**********************************************************************************************************************
 
-int ClassifySeqsCommand::driver(linePair* filePos, string taxFName, string tempTFName, string filename){
+int ClassifySeqsCommand::driver(linePair* filePos, string taxFName, string tempTFName, string accnos, string filename){
        try {
                ofstream outTax;
                m->openOutputFile(taxFName, outTax);
                
                ofstream outTaxSimple;
                m->openOutputFile(tempTFName, outTaxSimple);
+               
+               ofstream outAcc;
+               m->openOutputFile(accnos, outAcc);
        
                ifstream inFASTA;
                m->openInputFile(filename, inFASTA);
@@ -936,7 +982,11 @@ int ClassifySeqsCommand::driver(linePair* filePos, string taxFName, string tempT
                int count = 0;
                
                while (!done) {
-                       if (m->control_pressed) { return 0; }
+                       if (m->control_pressed) { 
+                               inFASTA.close();
+                               outTax.close();
+                               outTaxSimple.close();
+                               outAcc.close(); return 0; }
                
                        Sequence* candidateSeq = new Sequence(inFASTA); m->gobble(inFASTA);
                        
@@ -945,22 +995,25 @@ int ClassifySeqsCommand::driver(linePair* filePos, string taxFName, string tempT
                                taxonomy = classify->getTaxonomy(candidateSeq);
                                
                                if (m->control_pressed) { delete candidateSeq; return 0; }
-
-                               if (taxonomy != "bad seq") {
-                                       //output confidence scores or not
-                                       if (probs) {
-                                               outTax << candidateSeq->getName() << '\t' << taxonomy << endl;
-                                       }else{
-                                               outTax << candidateSeq->getName() << '\t' << classify->getSimpleTax() << endl;
-                                       }
-                                       
-                                       outTaxSimple << candidateSeq->getName() << '\t' << classify->getSimpleTax() << endl;
+                               
+                               if (taxonomy == "unknown;") { m->mothurOut("[WARNING]: " + candidateSeq->getName() + " could not be classified. You can use the remove.lineage command with taxon=unknown; to remove such sequences."); m->mothurOutEndLine(); }
+                               
+                               //output confidence scores or not
+                               if (probs) {
+                                       outTax << candidateSeq->getName() << '\t' << taxonomy << endl;
+                               }else{
+                                       outTax << candidateSeq->getName() << '\t' << classify->getSimpleTax() << endl;
                                }
+                               
+                               if (classify->getFlipped()) { outAcc << candidateSeq->getName() << endl; }
+                               
+                               outTaxSimple << candidateSeq->getName() << '\t' << classify->getSimpleTax() << endl;
+                               
                                count++;
                        }
                        delete candidateSeq;
                        
-                       #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
+                       #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) || (__linux__) || (__unix__) || (__unix)
                                unsigned long long pos = inFASTA.tellg();
                                if ((pos == -1) || (pos >= filePos->end)) { break; }
                        #else
@@ -977,6 +1030,7 @@ int ClassifySeqsCommand::driver(linePair* filePos, string taxFName, string tempT
                inFASTA.close();
                outTax.close();
                outTaxSimple.close();
+               outAcc.close();
                
                return count;
        }
@@ -987,10 +1041,11 @@ int ClassifySeqsCommand::driver(linePair* filePos, string taxFName, string tempT
 }
 //**********************************************************************************************************************
 #ifdef USE_MPI
-int ClassifySeqsCommand::driverMPI(int start, int num, MPI_File& inMPI, MPI_File& newFile, MPI_File& tempFile, vector<unsigned long long>& MPIPos){
+int ClassifySeqsCommand::driverMPI(int start, int num, MPI_File& inMPI, MPI_File& newFile, MPI_File& tempFile, MPI_File& accFile, vector<unsigned long long>& MPIPos){
        try {
                MPI_Status statusNew; 
                MPI_Status statusTemp; 
+               MPI_Status statusAcc; 
                MPI_Status status; 
                
                int pid;
@@ -1018,29 +1073,40 @@ int ClassifySeqsCommand::driverMPI(int start, int num, MPI_File& inMPI, MPI_File
                        if (candidateSeq->getName() != "") {
                                taxonomy = classify->getTaxonomy(candidateSeq);
                                
-                               if (taxonomy != "bad seq") {
-                                       //output confidence scores or not
-                                       if (probs) {
-                                               outputString =  candidateSeq->getName() + "\t" + taxonomy + "\n";
-                                       }else{
-                                               outputString =  candidateSeq->getName() + "\t" + classify->getSimpleTax() + "\n";
-                                       }
-                                       
-                                       int length = outputString.length();
-                                       char* buf2 = new char[length];
-                                       memcpy(buf2, outputString.c_str(), length);
+                               if (taxonomy == "unknown;") { m->mothurOut("[WARNING]: " + candidateSeq->getName() + " could not be classified. You can use the remove.lineage command with taxon=unknown; to remove such sequences."); m->mothurOutEndLine(); }
                                
-                                       MPI_File_write_shared(newFile, buf2, length, MPI_CHAR, &statusNew);
-                                       delete buf2;
-
+                               //output confidence scores or not
+                               if (probs) {
+                                       outputString =  candidateSeq->getName() + "\t" + taxonomy + "\n";
+                               }else{
                                        outputString =  candidateSeq->getName() + "\t" + classify->getSimpleTax() + "\n";
-                                       length = outputString.length();
-                                       char* buf = new char[length];
-                                       memcpy(buf, outputString.c_str(), length);
+                               }
+                               
+                               int length = outputString.length();
+                               char* buf2 = new char[length];
+                               memcpy(buf2, outputString.c_str(), length);
+                               
+                               MPI_File_write_shared(newFile, buf2, length, MPI_CHAR, &statusNew);
+                               delete buf2;
+                               
+                               outputString =  candidateSeq->getName() + "\t" + classify->getSimpleTax() + "\n";
+                               length = outputString.length();
+                               char* buf = new char[length];
+                               memcpy(buf, outputString.c_str(), length);
                                
-                                       MPI_File_write_shared(tempFile, buf, length, MPI_CHAR, &statusTemp);
-                                       delete buf;
+                               MPI_File_write_shared(tempFile, buf, length, MPI_CHAR, &statusTemp);
+                               delete buf;
+                               
+                               if (classify->getFlipped()) { 
+                                       outputString =  candidateSeq->getName() + "\n";
+                                       length = outputString.length();
+                                       char* buf3 = new char[length];
+                                       memcpy(buf3, outputString.c_str(), length);
+                                       
+                                       MPI_File_write_shared(accFile, buf3, length, MPI_CHAR, &statusAcc);
+                                       delete buf3;
                                }
+                               
                        }                               
                        delete candidateSeq;