X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=consensusseqscommand.cpp;h=d6158ba3ed3bc147027d54098aa71398182dc2f6;hb=078d7227da8dda9ae8620822fa32d51ec2706fc3;hp=bc47d8cca678ff9c8764651119ff255404a4ef65;hpb=ca9ac1d80c62f57270b0dcd49410ebe08a8aecd6;p=mothur.git diff --git a/consensusseqscommand.cpp b/consensusseqscommand.cpp index bc47d8c..d6158ba 100644 --- a/consensusseqscommand.cpp +++ b/consensusseqscommand.cpp @@ -18,6 +18,7 @@ vector ConsensusSeqsCommand::setParameters(){ CommandParameter pname("name", "InputTypes", "", "", "none", "none", "none",false,false); parameters.push_back(pname); CommandParameter plist("list", "InputTypes", "", "", "none", "none", "none",false,false); parameters.push_back(plist); CommandParameter plabel("label", "String", "", "", "", "", "",false,false); parameters.push_back(plabel); + CommandParameter pcutoff("cutoff", "Number", "", "100", "", "", "",false,false); parameters.push_back(pcutoff); CommandParameter pinputdir("inputdir", "String", "", "", "", "", "",false,false); parameters.push_back(pinputdir); CommandParameter poutputdir("outputdir", "String", "", "", "", "", "",false,false); parameters.push_back(poutputdir); @@ -35,15 +36,16 @@ string ConsensusSeqsCommand::getHelpString(){ try { string helpString = ""; helpString += "The consensus.seqs command can be used in 2 ways: create a consensus sequence from a fastafile, or with a listfile create a consensus sequence for each otu. Sequences must be aligned.\n"; - helpString += "The consensus.seqs command parameters are fasta, list, name and label.\n"; + helpString += "The consensus.seqs command parameters are fasta, list, name, cutoff and label.\n"; helpString += "The fasta parameter allows you to enter the fasta file containing your sequences, and is required, unless you have a valid current fasta file. \n"; helpString += "The list parameter allows you to enter a your list file. \n"; helpString += "The name parameter allows you to enter a names file associated with the fasta file. \n"; helpString += "The label parameter allows you to select what distance levels you would like output files for, and are separated by dashes.\n"; + helpString += "The cutoff parameter allows you set a percentage of sequences that support the base. For example: cutoff=97 would only return a sequence that only showed ambiguities for bases that were not supported by at least 97% of sequences.\n"; helpString += "The consensus.seqs command should be in the following format: \n"; helpString += "consensus.seqs(fasta=yourFastaFile, list=yourListFile) \n"; helpString += "Example: consensus.seqs(fasta=abrecovery.align, list=abrecovery.fn.list) \n"; - helpString += "Note: No spaces between parameter labels (i.e. fasta), '=' and parameters (i.e.yourFastaFile).\n\n"; + helpString += "Note: No spaces between parameter labels (i.e. fasta), '=' and parameters (i.e.yourFastaFile).\n"; return helpString; } catch(exception& e) { @@ -51,6 +53,28 @@ string ConsensusSeqsCommand::getHelpString(){ exit(1); } } +//********************************************************************************************************************** +string ConsensusSeqsCommand::getOutputFileNameTag(string type, string inputName=""){ + try { + string outputFileName = ""; + map >::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 == "fasta") { outputFileName = "cons.fasta"; } + else if (type == "name") { outputFileName = "cons.names"; } + else if (type == "summary") { outputFileName = "cons.summary"; } + 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, "ConsensusSeqsCommand", "getOutputFileNameTag"); + exit(1); + } +} //********************************************************************************************************************** ConsensusSeqsCommand::ConsensusSeqsCommand(){ @@ -75,6 +99,7 @@ ConsensusSeqsCommand::ConsensusSeqsCommand(string option) { //allow user to run help if(option == "help") { help(); abort = true; calledHelp = true; } + else if(option == "citation") { citation(); abort = true; calledHelp = true;} else { @@ -136,15 +161,17 @@ ConsensusSeqsCommand::ConsensusSeqsCommand(string option) { fastafile = m->getFastaFile(); if (fastafile != "") { m->mothurOut("Using " + fastafile + " as input file for the fasta parameter."); m->mothurOutEndLine(); } else { m->mothurOut("You have no current fastafile and the fasta parameter is required."); m->mothurOutEndLine(); abort = true; } - } + }else { m->setFastaFile(fastafile); } namefile = validParameter.validFile(parameters, "name", true); - if (namefile == "not open") { abort = true; } - else if (namefile == "not found") { namefile = ""; } + if (namefile == "not open") { namefile = ""; abort = true; } + else if (namefile == "not found") { namefile = ""; } + else { m->setNameFile(namefile); } listfile = validParameter.validFile(parameters, "list", true); if (listfile == "not open") { abort = true; } else if (listfile == "not found") { listfile = ""; } + else { m->setListFile(listfile); } label = validParameter.validFile(parameters, "label", false); if (label == "not found") { label = ""; } @@ -153,9 +180,16 @@ ConsensusSeqsCommand::ConsensusSeqsCommand(string option) { else { allLines = 1; } } + string temp = validParameter.validFile(parameters, "cutoff", false); if (temp == "not found") { temp = "100"; } + m->mothurConvert(temp, cutoff); + //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 = m->hasPath(fastafile); } - + + if (namefile == ""){ + vector files; files.push_back(fastafile); + parser.getNameFile(files); + } } } catch(exception& e) { @@ -182,7 +216,7 @@ int ConsensusSeqsCommand::execute(){ if (listfile == "") { ofstream outSummary; - string outputSummaryFile = outputDir + m->getRootName(m->getSimpleName(fastafile)) + "cons.summary"; + string outputSummaryFile = outputDir + m->getRootName(m->getSimpleName(fastafile)) + getOutputFileNameTag("summary"); m->openOutputFile(outputSummaryFile, outSummary); outSummary.setf(ios::fixed, ios::floatfield); outSummary.setf(ios::showpoint); outputNames.push_back(outputSummaryFile); outputTypes["summary"].push_back(outputSummaryFile); @@ -190,7 +224,7 @@ int ConsensusSeqsCommand::execute(){ outSummary << "PositioninAlignment\tA\tT\tG\tC\tGap\tNumberofSeqs\tConsensusBase" << endl; ofstream outFasta; - string outputFastaFile = outputDir + m->getRootName(m->getSimpleName(fastafile)) + "cons.fasta"; + string outputFastaFile = outputDir + m->getRootName(m->getSimpleName(fastafile)) + getOutputFileNameTag("fasta"); m->openOutputFile(outputFastaFile, outFasta); outputNames.push_back(outputFastaFile); outputTypes["fasta"].push_back(outputFastaFile); @@ -198,7 +232,7 @@ int ConsensusSeqsCommand::execute(){ int seqLength = 0; for (map::iterator it = nameMap.begin(); it != nameMap.end(); it++) { - if (m->control_pressed) { outSummary.close(); outFasta.close(); for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } return 0; } + if (m->control_pressed) { outSummary.close(); outFasta.close(); for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } return 0; } string seq = fastaMap[it->second]; seqs.push_back(seq); @@ -215,7 +249,7 @@ int ConsensusSeqsCommand::execute(){ //get counts for (int j = 0; j < seqLength; j++) { - if (m->control_pressed) { outSummary.close(); outFasta.close(); for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } return 0; } + if (m->control_pressed) { outSummary.close(); outFasta.close(); for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } return 0; } vector counts; counts.resize(5, 0); //A,T,G,C,Gap int numDots = 0; @@ -233,7 +267,7 @@ int ConsensusSeqsCommand::execute(){ } char conBase = '.'; - if (numDots != seqs.size()) { conBase = getBase(counts); } + if (numDots != seqs.size()) { conBase = getBase(counts, seqs.size()); } consSeq += conBase; @@ -268,7 +302,7 @@ int ConsensusSeqsCommand::execute(){ //as long as you are not at the end of the file or done wih the lines you want while((list != NULL) && ((allLines == 1) || (userLabels.size() != 0))) { - if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } delete list; delete input; return 0; } + if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } delete list; delete input; return 0; } if(allLines == 1 || labels.count(list->getLabel()) == 1){ @@ -306,7 +340,7 @@ int ConsensusSeqsCommand::execute(){ } - if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } if (list != NULL) { delete list; } delete input; return 0; } + if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } if (list != NULL) { delete list; } delete input; return 0; } //output error messages about any remaining user labels set::iterator it; @@ -358,18 +392,18 @@ int ConsensusSeqsCommand::processList(ListVector*& list){ try{ ofstream outSummary; - string outputSummaryFile = outputDir + m->getRootName(m->getSimpleName(fastafile)) + list->getLabel() + ".cons.summary"; + string outputSummaryFile = outputDir + m->getRootName(m->getSimpleName(fastafile)) + list->getLabel() + getOutputFileNameTag("summary"); m->openOutputFile(outputSummaryFile, outSummary); outSummary.setf(ios::fixed, ios::floatfield); outSummary.setf(ios::showpoint); outputNames.push_back(outputSummaryFile); outputTypes["summary"].push_back(outputSummaryFile); ofstream outName; - string outputNameFile = outputDir + m->getRootName(m->getSimpleName(fastafile)) + list->getLabel() + ".cons.names"; + string outputNameFile = outputDir + m->getRootName(m->getSimpleName(fastafile)) + list->getLabel() + getOutputFileNameTag("name"); m->openOutputFile(outputNameFile, outName); outputNames.push_back(outputNameFile); outputTypes["name"].push_back(outputNameFile); ofstream outFasta; - string outputFastaFile = outputDir + m->getRootName(m->getSimpleName(fastafile)) + list->getLabel() + ".cons.fasta"; + string outputFastaFile = outputDir + m->getRootName(m->getSimpleName(fastafile)) + list->getLabel() + getOutputFileNameTag("fasta"); m->openOutputFile(outputFastaFile, outFasta); outputNames.push_back(outputFastaFile); outputTypes["fasta"].push_back(outputFastaFile); @@ -476,7 +510,7 @@ string ConsensusSeqsCommand::getConsSeq(string bin, ofstream& outSummary, string } char conBase = '.'; - if (numDots != seqs.size()) { conBase = getBase(counts); } + if (numDots != seqs.size()) { conBase = getBase(counts, seqs.size()); } consSeq += conBase; @@ -502,7 +536,7 @@ string ConsensusSeqsCommand::getConsSeq(string bin, ofstream& outSummary, string } //*************************************************************************************************************** -char ConsensusSeqsCommand::getBase(vector counts){ //A,T,G,C,Gap +char ConsensusSeqsCommand::getBase(vector counts, int size){ //A,T,G,C,Gap try{ /* A = adenine * C = cytosine @@ -522,6 +556,14 @@ char ConsensusSeqsCommand::getBase(vector counts){ //A,T,G,C,Gap char conBase = 'N'; + //zero out counts that don't make the cutoff + float percentage = (100.0 - cutoff) / 100.0; + int zeroCutoff = percentage * size; + + for (int i = 0; i < counts.size(); i++) { + if (counts[i] < zeroCutoff) { counts[i] = 0; } + } + //any if ((counts[0] != 0) && (counts[1] != 0) && (counts[2] != 0) && (counts[3] != 0) && (counts[4] != 0)) { conBase = 'n'; } //any no gap @@ -584,6 +626,8 @@ char ConsensusSeqsCommand::getBase(vector counts){ //A,T,G,C,Gap else if ((counts[0] == 0) && (counts[1] == 0) && (counts[2] == 0) && (counts[3] != 0) && (counts[4] == 0)) { conBase = 'C'; } //only gap else if ((counts[0] == 0) && (counts[1] == 0) && (counts[2] == 0) && (counts[3] == 0) && (counts[4] != 0)) { conBase = '-'; } + //cutoff removed all counts + else if ((counts[0] == 0) && (counts[1] == 0) && (counts[2] == 0) && (counts[3] == 0) && (counts[4] == 0)) { conBase = 'N'; } else{ m->mothurOut("[ERROR]: cannot find consensus base."); m->mothurOutEndLine(); } return conBase; @@ -631,38 +675,29 @@ int ConsensusSeqsCommand::readFasta(){ int ConsensusSeqsCommand::readNames(){ try{ - - ifstream in; - m->openInputFile(namefile, in); - - string thisname, repnames; - map::iterator it; - - bool error = false; - - while(!in.eof()){ - - if (m->control_pressed) { break; } - - in >> thisname; m->gobble(in); //read from first column - in >> repnames; //read from second column - - it = nameMap.find(thisname); + map temp; + map::iterator it; + bool error = false; + + m->readNames(namefile, temp); //use central buffered read + + for (map::iterator itTemp = temp.begin(); itTemp != temp.end(); itTemp++) { + string thisname, repnames; + thisname = itTemp->first; + repnames = itTemp->second; + + it = nameMap.find(thisname); if (it != nameMap.end()) { //then this sequence was in the fastafile - + nameFileMap[thisname] = repnames; //for later when outputting the new namesFile if the list file is unique + vector splitRepNames; m->splitAtComma(repnames, splitRepNames); - nameFileMap[thisname] = repnames; //for later when outputting the new namesFile if the list file is unique for (int i = 0; i < splitRepNames.size(); i++) { nameMap[splitRepNames[i]] = thisname; } }else{ m->mothurOut("[ERROR]: " + thisname + " is not in the fasta file, please correct."); m->mothurOutEndLine(); error = true; } - - m->gobble(in); - } - - in.close(); - + } + if (error) { m->control_pressed = true; } return 0;