X-Git-Url: https://git.donarmstrong.com/?p=mothur.git;a=blobdiff_plain;f=consensusseqscommand.cpp;h=3d671f98420e0fa7cf844aac647bd43feea959e8;hp=84519b8e320817c63112d19d18be2f49c43e9228;hb=050a3ff02473a3d4c0980964e1a9ebe52e55d6b8;hpb=3fd6dd6e4f19a458ac2966ee5458787e998a1bde diff --git a/consensusseqscommand.cpp b/consensusseqscommand.cpp index 84519b8..3d671f9 100644 --- a/consensusseqscommand.cpp +++ b/consensusseqscommand.cpp @@ -12,68 +12,95 @@ #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","fasta-name",false,true,true); parameters.push_back(pfasta); + CommandParameter pname("name", "InputTypes", "", "", "namecount", "none", "none","name",false,false,true); parameters.push_back(pname); + CommandParameter pcount("count", "InputTypes", "", "", "namecount", "none", "none","count",false,false,true); parameters.push_back(pcount); + CommandParameter plist("list", "InputTypes", "", "", "none", "none", "none","fasta-name",false,false,true); 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"); + m->errorOut(e, "ConsensusSeqsCommand", "setParameters"); exit(1); } } //********************************************************************************************************************** -ConsensusSeqsCommand::ConsensusSeqsCommand(){ +string ConsensusSeqsCommand::getHelpString(){ try { - abort = true; - //initialize outputTypes - vector tempOutNames; - outputTypes["fasta"] = tempOutNames; - outputTypes["name"] = tempOutNames; - outputTypes["summary"] = tempOutNames; + 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, count, 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", "ConsensusSeqsCommand"); + m->errorOut(e, "ConsensusSeqsCommand", "getHelpString"); exit(1); } } //********************************************************************************************************************** -vector ConsensusSeqsCommand::getRequiredParameters(){ - try { - string Array[] = {"fasta", "list"}; - vector myArray (Array, Array+(sizeof(Array)/sizeof(string))); - return myArray; - } - catch(exception& e) { - m->errorOut(e, "ConsensusSeqsCommand", "getRequiredParameters"); - exit(1); - } +string ConsensusSeqsCommand::getOutputPattern(string type) { + try { + string pattern = ""; + + if (type == "fasta") { pattern = "[filename],cons.fasta-[filename],[tag],cons.fasta"; } + else if (type == "name") { pattern = "[filename],cons.names-[filename],[tag],cons.names"; } + else if (type == "count") { pattern = "[filename],cons.count_table-[filename],[tag],cons.count_table"; } + else if (type == "summary") { pattern = "[filename],cons.summary-[filename],[tag],cons.summary"; } + else { m->mothurOut("[ERROR]: No definition for type " + type + " output pattern.\n"); m->control_pressed = true; } + + return pattern; + } + catch(exception& e) { + m->errorOut(e, "ConsensusSeqsCommand", "getOutputPattern"); + 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["count"] = 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(); @@ -90,6 +117,7 @@ ConsensusSeqsCommand::ConsensusSeqsCommand(string option) { vector tempOutNames; outputTypes["fasta"] = tempOutNames; outputTypes["name"] = tempOutNames; + outputTypes["count"] = tempOutNames; outputTypes["summary"] = tempOutNames; @@ -121,21 +149,42 @@ ConsensusSeqsCommand::ConsensusSeqsCommand(string option) { //if the user has not given a path then, add inputdir. else leave path alone. if (path == "") { parameters["list"] = inputDir + it->second; } } + + it = parameters.find("count"); + //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["count"] = inputDir + it->second; } + } } //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); } + countfile = validParameter.validFile(parameters, "count", true); + if (countfile == "not open") { abort = true; countfile = ""; } + else if (countfile == "not found") { countfile = ""; } + else { m->setCountTableFile(countfile); } + + if ((countfile != "") && (namefile != "")) { m->mothurOut("You must enter ONLY ONE of the following: count or name."); m->mothurOutEndLine(); abort = true; } + 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 +193,18 @@ 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 (countfile == "") { + if (namefile == ""){ + vector files; files.push_back(fastafile); + parser.getNameFile(files); + } + } } } catch(exception& e) { @@ -154,127 +212,191 @@ 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; } + int start = time(NULL); + readFasta(); if (m->control_pressed) { return 0; } if (namefile != "") { readNames(); } + if (countfile != "") { ct.readTable(countfile, true); } 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 (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } delete list; delete input; return 0; } - - if(allLines == 1 || labels.count(list->getLabel()) == 1){ - m->mothurOut(list->getLabel()); m->mothurOutEndLine(); + if (listfile == "") { + + ofstream outSummary; + map variables; + variables["[filename]"] = outputDir + m->getRootName(m->getSimpleName(fastafile)); + string outputSummaryFile = getOutputFileName("summary", variables); + m->openOutputFile(outputSummaryFile, outSummary); + outSummary.setf(ios::fixed, ios::floatfield); outSummary.setf(ios::showpoint); + outputNames.push_back(outputSummaryFile); outputTypes["summary"].push_back(outputSummaryFile); + + outSummary << "PositioninAlignment\tA\tT\tG\tC\tGap\tNumberofSeqs\tConsensusBase" << endl; + + ofstream outFasta; + string outputFastaFile = getOutputFileName("fasta", variables); + m->openOutputFile(outputFastaFile, outFasta); + outputNames.push_back(outputFastaFile); outputTypes["fasta"].push_back(outputFastaFile); + + vector< vector > percentages; percentages.resize(5); + for (int j = 0; j < percentages.size(); j++) { percentages[j].resize(seqLength, 0.0); } + + string consSeq = ""; + int thisCount; + //get counts + for (int j = 0; j < seqLength; j++) { - processList(list); + 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; + thisCount = 0; + for (map::iterator it = fastaMap.begin(); it != fastaMap.end(); it++) { + + string thisSeq = it->second; + int size = 0; + + if (countfile != "") { size = ct.getNumSeqs(it->first); } + else { + map::iterator itCount = nameFileMap.find(it->first); + if (itCount != nameFileMap.end()) { + size = itCount->second; + }else { m->mothurOut("[ERROR]: file mismatch, aborting.\n"); m->control_pressed = true; break; } + } + + for (int k = 0; k < size; k++) { + if (thisSeq[j] == '.') { numDots++; } + + char base = toupper(thisSeq[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]++; } + thisCount++; + } + } - processedLabels.insert(list->getLabel()); - userLabels.erase(list->getLabel()); + char conBase = '.'; + if (numDots != thisCount) { conBase = getBase(counts, thisCount); } + + consSeq += conBase; + + percentages[0][j] = counts[0] / (float) thisCount; + percentages[1][j] = counts[1] / (float) thisCount; + percentages[2][j] = counts[2] / (float) thisCount; + percentages[3][j] = counts[3] / (float) thisCount; + percentages[4][j] = counts[4] / (float) thisCount; + } + + 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' << thisCount << '\t' << consSeq[j] << endl; } - if ((m->anyLabelsToProcess(list->getLabel(), userLabels, "") == true) && (processedLabels.count(lastLabel) != 1)) { - string saveLabel = list->getLabel(); - delete list; + outFasta << ">conseq" << endl << consSeq << endl; + + outSummary.close(); outFasta.close(); + + if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } return 0; } + + }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))) { - 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]); } delete list; delete input; return 0; } - processList(list); + 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(); - processedLabels.insert(list->getLabel()); - userLabels.erase(list->getLabel()); + delete list; list = NULL; - //restore real lastlabel to save below - list->setLabel(saveLabel); + //get next line to process + list = input->getListVector(); } - lastLabel = list->getLabel(); - delete list; list = NULL; + 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; } - //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(); + //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(); + } } - } - - //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); + //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->mothurOut("It took " + toString(time(NULL) - start) + " secs to find the consensus sequences."); + m->mothurOutEndLine(); - m->mothurOut("Output File Name: "); 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(); @@ -293,32 +415,44 @@ int ConsensusSeqsCommand::processList(ListVector*& list){ try{ ofstream outSummary; - string outputSummaryFile = outputDir + m->getRootName(m->getSimpleName(fastafile)) + list->getLabel() + ".cons.summary"; + map variables; + variables["[filename]"] = outputDir + m->getRootName(m->getSimpleName(fastafile)); + variables["[tag]"] = list->getLabel(); + string outputSummaryFile = getOutputFileName("summary", variables); 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 = getOutputFileName("name",variables); 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 = getOutputFileName("fasta",variables); m->openOutputFile(outputFastaFile, outFasta); outputNames.push_back(outputFastaFile); outputTypes["fasta"].push_back(outputFastaFile); + outSummary << "OTU#\tPositioninAlignment\tA\tT\tG\tC\tGap\tNumberofSeqs\tConsensusBase" << endl; + + string snumBins = toString(list->getNumBins()); for (int i = 0; i < list->getNumBins(); i++) { if (m->control_pressed) { outSummary.close(); outName.close(); outFasta.close(); return 0; } string bin = list->get(i); - - string newName = ""; - string consSeq = getConsSeq(bin, outSummary, newName, i); - - outFasta << ">seq" << (i+1) << endl << consSeq << endl; - outName << "seq" << (i+1) << '\t' << "seq" << (i+1) << "," << newName << endl; + string consSeq = getConsSeq(bin, outSummary, i); + + string seqName = "Otu"; + string sbinNumber = toString(i+1); + if (sbinNumber.length() < snumBins.length()) { + int diff = snumBins.length() - sbinNumber.length(); + for (int h = 0; h < diff; h++) { seqName += "0"; } + } + seqName += sbinNumber; + + outFasta << ">" << seqName << endl << consSeq << endl; + outName << seqName << '\t' << seqName << "," << bin << endl; } outSummary.close(); outName.close(); outFasta.close(); @@ -333,111 +467,128 @@ int ConsensusSeqsCommand::processList(ListVector*& list){ } //*************************************************************************************************************** -//made this smart enough to owrk with unique or non unique list file -string ConsensusSeqsCommand::getConsSeq(string bin, ofstream& outSummary, string& name, int binNumber){ +string ConsensusSeqsCommand::getConsSeq(string bin, ofstream& outSummary, int binNumber){ try{ string consSeq = ""; bool error = false; - - //the whole bin is the second column if no names file, otherwise build it - name = bin; - if (namefile != "") { name = ""; } - + int totalSize=0; + vector binNames; m->splitAtComma(bin, binNames); - - //get sequence strings for each name in the bin - vector seqs; - - set addedAlready; - int seqLength = 0; - for (int i = 0; i < binNames.size(); i++) { - - map::iterator it; - - it = nameMap.find(binNames[i]); - if (it == nameMap.end()) { - if (namefile == "") { m->mothurOut("[ERROR]: " + binNames[i] + " is not in your fasta file, please correct."); m->mothurOutEndLine(); error = true; } - else { m->mothurOut("[ERROR]: " + binNames[i] + " is not in your fasta or name file, please correct."); m->mothurOutEndLine(); error = true; } - break; - }else { - - //add sequence string to seqs vector to process below - string seq = fastaMap[it->second]; - seqs.push_back(seq); - - if (seqLength == 0) { seqLength = seq.length(); } - else if (seqLength != seq.length()) { m->mothurOut("[ERROR]: sequence are not the same length, please correct."); m->mothurOutEndLine(); error = true; break; } - - if (namefile != "") { - //did we add this line from name file already? - if (addedAlready.count(it->second) == 0) { - name += "," + nameFileMap[it->second]; - addedAlready.insert(it->second); - } - } - - } - } - - if (error) { m->control_pressed = true; return consSeq; } - - if (namefile != "") { name = name.substr(1); } - - vector< vector > percentages; percentages.resize(5); + + vector< vector > percentages; percentages.resize(5); for (int j = 0; j < percentages.size(); j++) { percentages[j].resize(seqLength, 0.0); } - - //get counts - for (int j = 0; j < seqLength; j++) { - - if (m->control_pressed) { return consSeq; } - - vector counts; counts.resize(5, 0); //A,T,G,C,Gap - int numDots = 0; - - 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); } - - 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(); - + + if (countfile != "") { + //get counts + for (int j = 0; j < seqLength; j++) { + + if (m->control_pressed) { return consSeq; } + + vector counts; counts.resize(5, 0); //A,T,G,C,Gap + int numDots = 0; + totalSize = 0; + for (int i = 0; i < binNames.size(); i++) { + if (m->control_pressed) { return consSeq; } + + string thisSeq = ""; + map::iterator itFasta = fastaMap.find(binNames[i]); + if (itFasta != fastaMap.end()) { + thisSeq = itFasta->second; + }else { m->mothurOut("[ERROR]: " + binNames[i] + " is not in your fasta file, please correct."); m->mothurOutEndLine(); m->control_pressed = true; } + + int size = ct.getNumSeqs(binNames[i]); + if (size != 0) { + for (int k = 0; k < size; k++) { + if (thisSeq[j] == '.') { numDots++; } + + char base = toupper(thisSeq[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]++; } + totalSize++; + } + }else { m->mothurOut("[ERROR]: " + binNames[i] + " is not in your count file, please correct."); m->mothurOutEndLine(); m->control_pressed = true; } + } + char conBase = '.'; + if (numDots != totalSize) { conBase = getBase(counts, totalSize); } + + consSeq += conBase; + + percentages[0][j] = counts[0] / (float) totalSize; + percentages[1][j] = counts[1] / (float) totalSize; + percentages[2][j] = counts[2] / (float) totalSize; + percentages[3][j] = counts[3] / (float) totalSize; + percentages[4][j] = counts[4] / (float) totalSize; + } + + }else { + + //get sequence strings for each name in the bin + vector seqs; + for (int i = 0; i < binNames.size(); i++) { + + map::iterator it; + it = nameMap.find(binNames[i]); + if (it == nameMap.end()) { + if (namefile == "") { m->mothurOut("[ERROR]: " + binNames[i] + " is not in your fasta file, please correct."); m->mothurOutEndLine(); error = true; } + else { m->mothurOut("[ERROR]: " + binNames[i] + " is not in your fasta or name file, please correct."); m->mothurOutEndLine(); error = true; } + break; + }else { + //add sequence string to seqs vector to process below + map::iterator itFasta = fastaMap.find(it->second); + + if (itFasta != fastaMap.end()) { + string seq = itFasta->second; + seqs.push_back(seq); + }else { m->mothurOut("[ERROR]: file mismatch, aborting. \n"); } + } + } + + if (error) { m->control_pressed = true; return consSeq; } + totalSize = seqs.size(); + //get counts + for (int j = 0; j < seqLength; j++) { + + if (m->control_pressed) { return consSeq; } + + vector counts; counts.resize(5, 0); //A,T,G,C,Gap + int numDots = 0; + + 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()); } + + 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(); + + } + } + + + + 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' << totalSize << '\t' << consSeq[j] << endl; } - - 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; - return consSeq; @@ -449,7 +600,7 @@ 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 +620,14 @@ 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; + + for (int i = 0; i < counts.size(); i++) { + float countPercentage = counts[i] / (float) size; + if (countPercentage < percentage) { 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 +690,8 @@ 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; @@ -549,7 +710,8 @@ int ConsensusSeqsCommand::readFasta(){ ifstream in; m->openInputFile(fastafile, in); - + seqLength = 0; + while (!in.eof()) { if (m->control_pressed) { break; } @@ -560,7 +722,10 @@ int ConsensusSeqsCommand::readFasta(){ if (name != "") { fastaMap[name] = seq.getAligned(); nameMap[name] = name; //set nameMap incase no names file - nameFileMap[name] = name; + nameFileMap[name] = 1; + + if (seqLength == 0) { seqLength = seq.getAligned().length(); } + else if (seqLength != seq.getAligned().length()) { m->mothurOut("[ERROR]: sequence are not the same length, please correct."); m->mothurOutEndLine(); m->control_pressed = true; break; } } } @@ -578,38 +743,29 @@ 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] = m->getNumNames(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;