X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=consensusseqscommand.cpp;h=55ec8023270cd17d38b830f519ad9f62518e635e;hb=ee8403d4eb5760187d62b42a9cf4272de8fc0ec4;hp=84519b8e320817c63112d19d18be2f49c43e9228;hpb=3fd6dd6e4f19a458ac2966ee5458787e998a1bde;p=mothur.git diff --git a/consensusseqscommand.cpp b/consensusseqscommand.cpp index 84519b8..55ec802 100644 --- a/consensusseqscommand.cpp +++ b/consensusseqscommand.cpp @@ -12,68 +12,76 @@ #include "inputdata.h" //********************************************************************************************************************** -vector ConsensusSeqsCommand::getValidParameters(){ +vector ConsensusSeqsCommand::setParameters(){ try { - string Array[] = {"fasta", "list", "name", "label","outputdir","inputdir"}; - vector myArray (Array, Array+(sizeof(Array)/sizeof(string))); + 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 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); + + vector myArray; + for (int i = 0; i < parameters.size(); i++) { myArray.push_back(parameters[i].name); } return myArray; } catch(exception& e) { - m->errorOut(e, "ConsensusSeqsCommand", "getValidParameters"); - exit(1); - } -} -//********************************************************************************************************************** -ConsensusSeqsCommand::ConsensusSeqsCommand(){ - try { - abort = true; - //initialize outputTypes - vector tempOutNames; - outputTypes["fasta"] = tempOutNames; - outputTypes["name"] = tempOutNames; - outputTypes["summary"] = tempOutNames; - } - catch(exception& e) { - m->errorOut(e, "ConsensusSeqsCommand", "ConsensusSeqsCommand"); + m->errorOut(e, "ConsensusSeqsCommand", "setParameters"); exit(1); } } //********************************************************************************************************************** -vector ConsensusSeqsCommand::getRequiredParameters(){ +string ConsensusSeqsCommand::getHelpString(){ try { - string Array[] = {"fasta", "list"}; - vector myArray (Array, Array+(sizeof(Array)/sizeof(string))); - return myArray; + 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, 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"; + return helpString; } catch(exception& e) { - m->errorOut(e, "ConsensusSeqsCommand", "getRequiredParameters"); + m->errorOut(e, "ConsensusSeqsCommand", "getHelpString"); exit(1); } } + //********************************************************************************************************************** -vector ConsensusSeqsCommand::getRequiredFiles(){ +ConsensusSeqsCommand::ConsensusSeqsCommand(){ try { - vector myArray; - return myArray; + abort = true; calledHelp = true; + setParameters(); + vector tempOutNames; + outputTypes["fasta"] = tempOutNames; + outputTypes["name"] = tempOutNames; + outputTypes["summary"] = tempOutNames; } catch(exception& e) { - m->errorOut(e, "DegapSeqsCommand", "getRequiredFiles"); + m->errorOut(e, "ConsensusSeqsCommand", "ConsensusSeqsCommand"); exit(1); } } //*************************************************************************************************************** ConsensusSeqsCommand::ConsensusSeqsCommand(string option) { try { - abort = false; + abort = false; calledHelp = false; allLines = 1; //allow user to run help - if(option == "help") { help(); abort = true; } + if(option == "help") { help(); abort = true; calledHelp = true; } + else if(option == "citation") { citation(); abort = true; calledHelp = true;} else { - //valid paramters for this command - string Array[] = {"fasta","list","name","label", "outputdir","inputdir"}; - vector myArray (Array, Array+(sizeof(Array)/sizeof(string))); + + vector myArray = setParameters(); OptionParser parser(option); map parameters = parser.getParameters(); @@ -127,15 +135,21 @@ ConsensusSeqsCommand::ConsensusSeqsCommand(string option) { //check for parameters fastafile = validParameter.validFile(parameters, "fasta", true); if (fastafile == "not open") { abort = true; } - else if (fastafile == "not found") { fastafile = ""; m->mothurOut("fasta is a required parameter for the consensus.seqs command."); m->mothurOutEndLine(); abort = true; } + else if (fastafile == "not found") { + 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 = ""; m->mothurOut("list is a required parameter for the consensus.seqs command."); m->mothurOutEndLine(); abort = true; } + else if (listfile == "not found") { listfile = ""; } + else { m->setListFile(listfile); } label = validParameter.validFile(parameters, "label", false); if (label == "not found") { label = ""; } @@ -144,9 +158,16 @@ 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) { @@ -154,37 +175,12 @@ ConsensusSeqsCommand::ConsensusSeqsCommand(string option) { exit(1); } } -//********************************************************************************************************************** - -void ConsensusSeqsCommand::help(){ - try { - m->mothurOut("The consensus.seqs command reads a fastafile and listfile and creates a consensus sequence for each otu. Sequences must be aligned.\n"); - m->mothurOut("The consensus.seqs command parameters are fasta, list, name and label.\n"); - m->mothurOut("The fasta parameter allows you to enter the fasta file containing your sequences, and is required. \n"); - m->mothurOut("The list parameter allows you to enter a your list file. \n"); - m->mothurOut("The name parameter allows you to enter a names file associated with the fasta file. \n"); - m->mothurOut("The label parameter allows you to select what distance levels you would like output files for, and are separated by dashes.\n"); - m->mothurOut("The consensus.seqs command should be in the following format: \n"); - m->mothurOut("consensus.seqs(fasta=yourFastaFile, list=yourListFile) \n"); - m->mothurOut("Example: consensus.seqs(fasta=abrecovery.align, list=abrecovery.fn.list) \n"); - m->mothurOut("Note: No spaces between parameter labels (i.e. fasta), '=' and parameters (i.e.yourFastaFile).\n\n"); - } - catch(exception& e) { - m->errorOut(e, "ConsensusSeqsCommand", "help"); - exit(1); - } -} - -//*************************************************************************************************************** - -ConsensusSeqsCommand::~ConsensusSeqsCommand(){ /* do nothing */ } - //*************************************************************************************************************** int ConsensusSeqsCommand::execute(){ try{ - if (abort == true) { return 0; } + if (abort == true) { if (calledHelp) { return 0; } return 2; } readFasta(); @@ -194,85 +190,166 @@ int ConsensusSeqsCommand::execute(){ if (m->control_pressed) { return 0; } - InputData* input = new InputData(listfile, "list"); - ListVector* list = input->getListVector(); - - string lastLabel = list->getLabel(); - set processedLabels; - set userLabels = labels; - - //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 (listfile == "") { + + ofstream outSummary; + string outputSummaryFile = outputDir + m->getRootName(m->getSimpleName(fastafile)) + "cons.summary"; + m->openOutputFile(outputSummaryFile, outSummary); + outSummary.setf(ios::fixed, ios::floatfield); outSummary.setf(ios::showpoint); + outputNames.push_back(outputSummaryFile); outputTypes["summary"].push_back(outputSummaryFile); - if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } delete list; delete input; return 0; } + outSummary << "PositioninAlignment\tA\tT\tG\tC\tGap\tNumberofSeqs\tConsensusBase" << endl; - if(allLines == 1 || labels.count(list->getLabel()) == 1){ + ofstream outFasta; + string outputFastaFile = outputDir + m->getRootName(m->getSimpleName(fastafile)) + "cons.fasta"; + m->openOutputFile(outputFastaFile, outFasta); + outputNames.push_back(outputFastaFile); outputTypes["fasta"].push_back(outputFastaFile); + + vector seqs; + int seqLength = 0; + for (map::iterator it = nameMap.begin(); it != nameMap.end(); it++) { - m->mothurOut(list->getLabel()); m->mothurOutEndLine(); + if (m->control_pressed) { outSummary.close(); outFasta.close(); for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } return 0; } - processList(list); + string seq = fastaMap[it->second]; + seqs.push_back(seq); - processedLabels.insert(list->getLabel()); - userLabels.erase(list->getLabel()); + if (seqLength == 0) { seqLength = seq.length(); } + else if (seqLength != seq.length()) { m->mothurOut("[ERROR]: sequence are not the same length, please correct."); m->mothurOutEndLine(); m->control_pressed = true; } + } - if ((m->anyLabelsToProcess(list->getLabel(), userLabels, "") == true) && (processedLabels.count(lastLabel) != 1)) { - string saveLabel = list->getLabel(); + vector< vector > percentages; percentages.resize(5); + for (int j = 0; j < percentages.size(); j++) { percentages[j].resize(seqLength, 0.0); } + + string consSeq = ""; + //get counts + for (int j = 0; j < seqLength; j++) { - delete list; + if (m->control_pressed) { outSummary.close(); outFasta.close(); for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } return 0; } - list = input->getListVector(lastLabel); - m->mothurOut(list->getLabel()); m->mothurOutEndLine(); + vector counts; counts.resize(5, 0); //A,T,G,C,Gap + int numDots = 0; - processList(list); + for (int i = 0; i < seqs.size(); i++) { + + if (seqs[i][j] == '.') { numDots++; } + + char base = toupper(seqs[i][j]); + if (base == 'A') { counts[0]++; } + else if (base == 'T') { counts[1]++; } + else if (base == 'G') { counts[2]++; } + else if (base == 'C') { counts[3]++; } + else { counts[4]++; } + } + + char conBase = '.'; + if (numDots != seqs.size()) { conBase = getBase(counts, seqs.size()); } - processedLabels.insert(list->getLabel()); - userLabels.erase(list->getLabel()); + consSeq += conBase; + + percentages[0][j] = counts[0] / (float) seqs.size(); + percentages[1][j] = counts[1] / (float) seqs.size(); + percentages[2][j] = counts[2] / (float) seqs.size(); + percentages[3][j] = counts[3] / (float) seqs.size(); + percentages[4][j] = counts[4] / (float) seqs.size(); - //restore real lastlabel to save below - list->setLabel(saveLabel); } - lastLabel = list->getLabel(); + for (int j = 0; j < seqLength; j++) { + outSummary << (j+1) << '\t' << percentages[0][j] << '\t'<< percentages[1][j] << '\t'<< percentages[2][j] << '\t' << percentages[3][j] << '\t' << percentages[4][j] << '\t' << seqs.size() << '\t' << consSeq[j] << endl; + } - delete list; list = NULL; + + outFasta << ">conseq" << endl << consSeq << endl; + + outSummary.close(); outFasta.close(); - //get next line to process - list = input->getListVector(); - } - - - 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; } - //output error messages about any remaining user labels - set::iterator it; - bool needToRun = false; - for (it = userLabels.begin(); it != userLabels.end(); it++) { - m->mothurOut("Your file does not include the label " + *it); - if (processedLabels.count(lastLabel) != 1) { - m->mothurOut(". I will use " + lastLabel + "."); m->mothurOutEndLine(); - needToRun = true; - }else { - m->mothurOut(". Please refer to " + lastLabel + "."); m->mothurOutEndLine(); + }else { + + + InputData* input = new InputData(listfile, "list"); + ListVector* list = input->getListVector(); + + string lastLabel = list->getLabel(); + set processedLabels; + set userLabels = labels; + + //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++) { m->mothurRemove(outputNames[i]); } delete list; delete input; return 0; } + + if(allLines == 1 || labels.count(list->getLabel()) == 1){ + + m->mothurOut(list->getLabel()); m->mothurOutEndLine(); + + processList(list); + + processedLabels.insert(list->getLabel()); + userLabels.erase(list->getLabel()); + } + + if ((m->anyLabelsToProcess(list->getLabel(), userLabels, "") == true) && (processedLabels.count(lastLabel) != 1)) { + string saveLabel = list->getLabel(); + + delete list; + + list = input->getListVector(lastLabel); + m->mothurOut(list->getLabel()); m->mothurOutEndLine(); + + processList(list); + + processedLabels.insert(list->getLabel()); + userLabels.erase(list->getLabel()); + + //restore real lastlabel to save below + list->setLabel(saveLabel); + } + + lastLabel = list->getLabel(); + + delete list; list = NULL; + + //get next line to process + list = input->getListVector(); } - } - - //run last label if you need to - if (needToRun == true) { - if (list != NULL) { delete list; } - list = input->getListVector(lastLabel); - m->mothurOut(list->getLabel()); m->mothurOutEndLine(); + 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; + bool needToRun = false; + for (it = userLabels.begin(); it != userLabels.end(); it++) { + m->mothurOut("Your file does not include the label " + *it); + if (processedLabels.count(lastLabel) != 1) { + m->mothurOut(". I will use " + lastLabel + "."); m->mothurOutEndLine(); + needToRun = true; + }else { + m->mothurOut(". Please refer to " + lastLabel + "."); m->mothurOutEndLine(); + } + } - processList(list); + //run last label if you need to + if (needToRun == true) { + if (list != NULL) { delete list; } + + list = input->getListVector(lastLabel); + + m->mothurOut(list->getLabel()); m->mothurOutEndLine(); + + processList(list); + + delete list; list = NULL; + } - delete list; list = NULL; + if (list != NULL) { delete list; } + delete input; } - if (list != NULL) { delete list; } - delete input; - m->mothurOutEndLine(); m->mothurOut("Output File Name: "); m->mothurOutEndLine(); for (int i = 0; i < outputNames.size(); i++) { m->mothurOut(outputNames[i]); m->mothurOutEndLine(); } @@ -308,6 +385,8 @@ int ConsensusSeqsCommand::processList(ListVector*& list){ m->openOutputFile(outputFastaFile, outFasta); outputNames.push_back(outputFastaFile); outputTypes["fasta"].push_back(outputFastaFile); + outSummary << "OTU#\tPositioninAlignment\tA\tT\tG\tC\tGap\tNumberofSeqs\tConsensusBase" << endl; + for (int i = 0; i < list->getNumBins(); i++) { if (m->control_pressed) { outSummary.close(); outName.close(); outFasta.close(); return 0; } @@ -409,7 +488,7 @@ 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; @@ -421,23 +500,9 @@ string ConsensusSeqsCommand::getConsSeq(string bin, ofstream& outSummary, string } - outSummary << ">seq" << (binNumber + 1) << endl; - outSummary << "A" << '\t'; - for (int j = 0; j < seqLength; j++) { outSummary << percentages[0][j] << '\t'; } - outSummary << endl; - outSummary << "T" << '\t'; - for (int j = 0; j < seqLength; j++) { outSummary << percentages[1][j] << '\t'; } - outSummary << endl; - outSummary << "G" << '\t'; - for (int j = 0; j < seqLength; j++) { outSummary << percentages[2][j] << '\t'; } - outSummary << endl; - outSummary << "C" << '\t'; - for (int j = 0; j < seqLength; j++) { outSummary << percentages[3][j] << '\t'; } - outSummary << endl; - outSummary << "Gap" << '\t'; - for (int j = 0; j < seqLength; j++) { outSummary << percentages[4][j] << '\t'; } - outSummary << endl; - + for (int j = 0; j < seqLength; j++) { + outSummary << (binNumber + 1) << '\t' << (j+1) << '\t' << percentages[0][j] << '\t'<< percentages[1][j] << '\t'<< percentages[2][j] << '\t' << percentages[3][j] << '\t' << percentages[4][j] << '\t' << seqs.size() << '\t' << consSeq[j] << endl; + } return consSeq; @@ -449,7 +514,7 @@ 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 @@ -469,6 +534,14 @@ 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 @@ -531,6 +604,8 @@ 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; @@ -578,38 +653,29 @@ 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;