X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=trimseqscommand.cpp;h=b430624838594e4185233f7e946733a94469303b;hb=bbdb72971ea4cd171abe34985af89f97bc2a31d3;hp=ef7dfdd0f45d649cc6296d91e106e75b9a6fd167;hpb=c9c6df74a855bc1fcecb93cf56eb09d0907e7cf2;p=mothur.git diff --git a/trimseqscommand.cpp b/trimseqscommand.cpp index ef7dfdd..b430624 100644 --- a/trimseqscommand.cpp +++ b/trimseqscommand.cpp @@ -8,6 +8,7 @@ */ #include "trimseqscommand.h" +#include "needlemanoverlap.hpp" //*************************************************************************************************************** @@ -15,6 +16,7 @@ TrimSeqsCommand::TrimSeqsCommand(string option) { try { abort = false; + comboStarts = 0; //allow user to run help if(option == "help") { help(); abort = true; } @@ -22,7 +24,7 @@ TrimSeqsCommand::TrimSeqsCommand(string option) { else { //valid paramters for this command string AlignArray[] = {"fasta", "flip", "oligos", "maxambig", "maxhomop", "minlength", "maxlength", "qfile", - "qthreshold", "qaverage", "allfiles", "qtrim", "processors", "outputdir","inputdir"}; + "qthreshold", "qwindowaverage", "qstepsize", "qwindowsize", "qaverage", "rollaverage", "allfiles", "qtrim","tdiffs", "pdiffs", "bdiffs", "processors", "outputdir","inputdir"}; vector myArray (AlignArray, AlignArray+(sizeof(AlignArray)/sizeof(string))); @@ -79,6 +81,7 @@ TrimSeqsCommand::TrimSeqsCommand(string option) { outputDir += hasPath(fastaFile); //if user entered a file with a path then preserve it } + //check for optional parameter and set defaults // ...at some point should added some additional type checking... string temp; @@ -103,6 +106,17 @@ TrimSeqsCommand::TrimSeqsCommand(string option) { temp = validParameter.validFile(parameters, "maxlength", false); if (temp == "not found") { temp = "0"; } convert(temp, maxLength); + temp = validParameter.validFile(parameters, "bdiffs", false); if (temp == "not found") { temp = "0"; } + convert(temp, bdiffs); + + temp = validParameter.validFile(parameters, "pdiffs", false); if (temp == "not found") { temp = "0"; } + convert(temp, pdiffs); + + temp = validParameter.validFile(parameters, "tdiffs", false); if (temp == "not found") { int tempTotal = pdiffs + bdiffs; temp = toString(tempTotal); } + convert(temp, tdiffs); + + if(tdiffs == 0){ tdiffs = bdiffs + pdiffs; } + temp = validParameter.validFile(parameters, "qfile", true); if (temp == "not found") { qFileName = ""; } else if(temp == "not open") { abort = true; } @@ -111,16 +125,28 @@ TrimSeqsCommand::TrimSeqsCommand(string option) { temp = validParameter.validFile(parameters, "qthreshold", false); if (temp == "not found") { temp = "0"; } convert(temp, qThreshold); - temp = validParameter.validFile(parameters, "qtrim", false); if (temp == "not found") { temp = "F"; } + temp = validParameter.validFile(parameters, "qtrim", false); if (temp == "not found") { temp = "F"; } qtrim = isTrue(temp); + temp = validParameter.validFile(parameters, "rollaverage", false); if (temp == "not found") { temp = "0"; } + convert(temp, qRollAverage); + + temp = validParameter.validFile(parameters, "qwindowaverage", false);if (temp == "not found") { temp = "0"; } + convert(temp, qWindowAverage); + + temp = validParameter.validFile(parameters, "qwindowsize", false); if (temp == "not found") { temp = "100"; } + convert(temp, qWindowSize); + + temp = validParameter.validFile(parameters, "qstepsize", false); if (temp == "not found") { temp = "10"; } + convert(temp, qWindowStep); + temp = validParameter.validFile(parameters, "qaverage", false); if (temp == "not found") { temp = "0"; } convert(temp, qAverage); temp = validParameter.validFile(parameters, "allfiles", false); if (temp == "not found") { temp = "F"; } allFiles = isTrue(temp); - temp = validParameter.validFile(parameters, "processors", false); if (temp == "not found"){ temp = "1"; } + temp = validParameter.validFile(parameters, "processors", false); if (temp == "not found") { temp = "1"; } convert(temp, processors); if(allFiles && oligoFile == ""){ @@ -148,14 +174,17 @@ TrimSeqsCommand::TrimSeqsCommand(string option) { void TrimSeqsCommand::help(){ try { m->mothurOut("The trim.seqs command reads a fastaFile and creates .....\n"); - m->mothurOut("The trim.seqs command parameters are fasta, flip, oligos, maxambig, maxhomop, minlength, maxlength, qfile, qthreshold, qaverage, qtrim and allfiles.\n"); + m->mothurOut("The trim.seqs command parameters are fasta, flip, oligos, maxambig, maxhomop, minlength, maxlength, qfile, qthreshold, qaverage, diffs, qtrim and allfiles.\n"); m->mothurOut("The fasta parameter is required.\n"); - m->mothurOut("The flip parameter .... The default is 0.\n"); + m->mothurOut("The flip parameter will output the reverse compliment of your trimmed sequence. The default is false.\n"); m->mothurOut("The oligos parameter .... The default is "".\n"); m->mothurOut("The maxambig parameter .... The default is -1.\n"); m->mothurOut("The maxhomop parameter .... The default is 0.\n"); m->mothurOut("The minlength parameter .... The default is 0.\n"); m->mothurOut("The maxlength parameter .... The default is 0.\n"); + m->mothurOut("The tdiffs parameter is used to specify the total number of differences allowed in the sequence. The default is pdiffs + bdiffs.\n"); + m->mothurOut("The bdiffs parameter is used to specify the number of differences allowed in the barcode. The default is 0.\n"); + m->mothurOut("The pdiffs parameter is used to specify the number of differences allowed in the primer. The default is 0.\n"); m->mothurOut("The qfile parameter .....\n"); m->mothurOut("The qthreshold parameter .... The default is 0.\n"); m->mothurOut("The qaverage parameter .... The default is 0.\n"); @@ -194,12 +223,17 @@ int TrimSeqsCommand::execute(){ outputNames.push_back(trimSeqFile); string scrapSeqFile = outputDir + getRootName(getSimpleName(fastaFile)) + "scrap.fasta"; outputNames.push_back(scrapSeqFile); + string trimQualFile = outputDir + getRootName(getSimpleName(fastaFile)) + "trim.qual"; + outputNames.push_back(trimQualFile); + string scrapQualFile = outputDir + getRootName(getSimpleName(fastaFile)) + "scrap.qual"; + outputNames.push_back(scrapQualFile); string groupFile = outputDir + getRootName(getSimpleName(fastaFile)) + "groups"; vector fastaFileNames; + vector qualFileNames; if(oligoFile != ""){ outputNames.push_back(groupFile); - getOligos(fastaFileNames); + getOligos(fastaFileNames, qualFileNames); } if(qFileName != "") { setLines(qFileName, qLines); } @@ -208,79 +242,161 @@ int TrimSeqsCommand::execute(){ #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) if(processors == 1){ ifstream inFASTA; + int numSeqs; openInputFile(fastaFile, inFASTA); - int numSeqs=count(istreambuf_iterator(inFASTA),istreambuf_iterator(), '>'); + getNumSeqs(inFASTA, numSeqs); inFASTA.close(); lines.push_back(new linePair(0, numSeqs)); - driverCreateTrim(fastaFile, qFileName, trimSeqFile, scrapSeqFile, groupFile, fastaFileNames, lines[0], lines[0]); + driverCreateTrim(fastaFile, qFileName, trimSeqFile, scrapSeqFile, trimQualFile, scrapQualFile, groupFile, fastaFileNames, qualFileNames, lines[0], lines[0]); for (int j = 0; j < fastaFileNames.size(); j++) { rename((fastaFileNames[j] + toString(getpid()) + ".temp").c_str(), fastaFileNames[j].c_str()); } + if(qFileName != ""){ + for (int j = 0; j < qualFileNames.size(); j++) { + rename((qualFileNames[j] + toString(getpid()) + ".temp").c_str(), qualFileNames[j].c_str()); + } + } }else{ setLines(fastaFile, lines); if(qFileName == "") { qLines = lines; } - createProcessesCreateTrim(fastaFile, qFileName, trimSeqFile, scrapSeqFile, groupFile, fastaFileNames); + createProcessesCreateTrim(fastaFile, qFileName, trimSeqFile, scrapSeqFile, trimQualFile, scrapQualFile, groupFile, fastaFileNames, qualFileNames); rename((trimSeqFile + toString(processIDS[0]) + ".temp").c_str(), trimSeqFile.c_str()); rename((scrapSeqFile + toString(processIDS[0]) + ".temp").c_str(), scrapSeqFile.c_str()); rename((groupFile + toString(processIDS[0]) + ".temp").c_str(), groupFile.c_str()); + + if(qFileName != ""){ + rename((trimQualFile + toString(processIDS[0]) + ".temp").c_str(), trimQualFile.c_str()); + rename((scrapQualFile + toString(processIDS[0]) + ".temp").c_str(), scrapQualFile.c_str()); + } + + for (int j = 0; j < fastaFileNames.size(); j++) { rename((fastaFileNames[j] + toString(processIDS[0]) + ".temp").c_str(), fastaFileNames[j].c_str()); } + if(qFileName != ""){ + for (int j = 0; j < qualFileNames.size(); j++) { + rename((qualFileNames[j] + toString(getpid()) + ".temp").c_str(), qualFileNames[j].c_str()); + } + } + //append files for(int i=1;icontrol_pressed) { return 0; } #else ifstream inFASTA; - openInputFile(fastafileNames[s], inFASTA); - numSeqs=count(istreambuf_iterator(inFASTA),istreambuf_iterator(), '>'); + int numSeqs; + openInputFile(fastaFile, inFASTA); + getNumSeqs(inFASTA, numSeqs); inFASTA.close(); lines.push_back(new linePair(0, numSeqs)); - driverCreateSummary(fastafile, qFileName, trimSeqFile, scrapSeqFile, groupFile, fastaFileNames, lines[0], lines[0]); + driverCreateTrim(fastaFile, qFileName, trimSeqFile, scrapSeqFile, trimQualFile, scrapQualFile, groupFile, fastaFileNames, qualFileNames, lines[0], lines[0]); + for (int j = 0; j < fastaFileNames.size(); j++) { + rename((fastaFileNames[j] + toString(j) + ".temp").c_str(), fastaFileNames[j].c_str()); + } + if(qFileName != ""){ + for (int j = 0; j < qualFileNames.size(); j++) { + rename((qualFileNames[j] + toString(j) + ".temp").c_str(), qualFileNames[j].c_str()); + } + } + if (m->control_pressed) { return 0; } #endif for(int i=0;i'){ - inFASTA >> seqName; - outGroups << seqName << '\t' << groupVector[i] << endl; - } - while (!inFASTA.eof()) { char c = inFASTA.get(); if (c == 10 || c == 13){ break; } } + if (isBlank(fastaFileNames[i])) { remove(fastaFileNames[i].c_str()); } + else if (filesToRemove.count(fastaFileNames[i]) > 0) { remove(fastaFileNames[i].c_str()); } + else { + ifstream inFASTA; + string seqName; + openInputFile(fastaFileNames[i], inFASTA); + ofstream outGroups; + string outGroupFilename = outputDir + getRootName(getSimpleName(fastaFileNames[i])) + "groups"; + openOutputFile(outGroupFilename, outGroups); + outputNames.push_back(outGroupFilename); + + string thisGroup = ""; + if (i > comboStarts) { + map::iterator itCombo; + for(itCombo=combos.begin();itCombo!=combos.end(); itCombo++){ + if(itCombo->second == i){ thisGroup = itCombo->first; combos.erase(itCombo); break; } + } + }else{ thisGroup = groupVector[i]; } + + while(!inFASTA.eof()){ + if(inFASTA.get() == '>'){ + inFASTA >> seqName; + outGroups << seqName << '\t' << thisGroup << endl; + } + while (!inFASTA.eof()) { char c = inFASTA.get(); if (c == 10 || c == 13){ break; } } + } + outGroups.close(); + inFASTA.close(); + } + } + + if(qFileName != ""){ + for(int i=0;i 0) { remove(qualFileNames[i].c_str()); } + else { + ifstream inQual; + string seqName; + openInputFile(qualFileNames[i], inQual); +// ofstream outGroups; +// +// string thisGroup = ""; +// if (i > comboStarts) { +// map::iterator itCombo; +// for(itCombo=combos.begin();itCombo!=combos.end(); itCombo++){ +// if(itCombo->second == i){ thisGroup = itCombo->first; combos.erase(itCombo); break; } +// } +// } +// else{ thisGroup = groupVector[i]; } + + inQual.close(); + } } - outGroups.close(); - inFASTA.close(); } + if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } return 0; @@ -301,7 +417,9 @@ int TrimSeqsCommand::execute(){ } /**************************************************************************************/ -int TrimSeqsCommand::driverCreateTrim(string filename, string qFileName, string trimFile, string scrapFile, string groupFile, vector fastaNames, linePair* line, linePair* qline) { + +int TrimSeqsCommand::driverCreateTrim(string filename, string qFileName, string trimFile, string scrapFile, string trimQFile, string scrapQFile, string groupFile, vector fastaNames, vector qualNames, linePair* line, linePair* qline) { + try { ofstream outFASTA; @@ -310,12 +428,32 @@ int TrimSeqsCommand::driverCreateTrim(string filename, string qFileName, string ofstream scrapFASTA; openOutputFile(scrapFile, scrapFASTA); + ofstream outQual; + ofstream scrapQual; + if(qFileName != ""){ + openOutputFile(trimQFile, outQual); + openOutputFile(scrapQFile, scrapQual); + } + ofstream outGroups; vector fastaFileNames; + vector qualFileNames; + + if (oligoFile != "") { openOutputFile(groupFile, outGroups); for (int i = 0; i < fastaNames.size(); i++) { + #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) fastaFileNames.push_back(new ofstream((fastaNames[i] + toString(getpid()) + ".temp").c_str(), ios::ate)); + if(qFileName != ""){ + qualFileNames.push_back(new ofstream((qualNames[i] + toString(getpid()) + ".temp").c_str(), ios::ate)); + } + #else + fastaFileNames.push_back(new ofstream((fastaNames[i] + toString(i) + ".temp").c_str(), ios::ate)); + if(qFileName != ""){ + qualFileNames.push_back(new ofstream((qualNames[i] + toString(i) + ".temp").c_str(), ios::ate)); + } + #endif } } @@ -331,98 +469,136 @@ int TrimSeqsCommand::driverCreateTrim(string filename, string qFileName, string for(int i=0;inum;i++){ if (m->control_pressed) { - inFASTA.close(); - outFASTA.close(); - scrapFASTA.close(); + inFASTA.close(); outFASTA.close(); scrapFASTA.close(); if (oligoFile != "") { outGroups.close(); } - if(qFileName != "") { qFile.close(); } - for(int i=0;iclose(); - delete fastaFileNames[i]; - } + + for(int i=0;iclose(); delete fastaFileNames[i]; } + + if(qFileName != ""){ + qFile.close(); + for(int i=0;iclose(); delete qualFileNames[i]; } + } for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } + return 0; } - bool success = 1; + int success = 1; Sequence currSeq(inFASTA); - + QualityScores currQual; + if(qFileName != ""){ + currQual = QualityScores(qFile, currSeq.getNumBases()); + } + string origSeq = currSeq.getUnaligned(); if (origSeq != "") { - int group; + int groupBar, groupPrime; string trashCode = ""; - + int currentSeqsDiffs = 0; + if(qFileName != ""){ - if(qThreshold != 0) { success = stripQualThreshold(currSeq, qFile); } - else if(qAverage != 0) { success = cullQualAverage(currSeq, qFile); } - if ((!qtrim) && (origSeq.length() != currSeq.getUnaligned().length())) { + if(qThreshold != 0) { success = currQual.stripQualThreshold(currSeq, qThreshold); } + else if(qAverage != 0) { success = currQual.cullQualAverage(currSeq, qAverage); } + else if(qRollAverage != 0) { success = currQual.stripQualRollingAverage(currSeq, qRollAverage); } + else if(qWindowAverage != 0){ success = currQual.stripQualWindowAverage(currSeq, qWindowStep, qWindowSize, qWindowAverage); } + + if (qtrim == 1 && (origSeq.length() != currSeq.getUnaligned().length())) { success = 0; //if you don't want to trim and the sequence does not meet quality requirements, move to scrap } - if(!success) { trashCode += 'q'; } + + if(!success) { trashCode += 'q'; } } if(barcodes.size() != 0){ - success = stripBarcode(currSeq, group); - if(!success){ trashCode += 'b'; } + success = stripBarcode(currSeq, currQual, groupBar); + if(success > bdiffs) { trashCode += 'b'; } + else{ currentSeqsDiffs += success; } } - + if(numFPrimers != 0){ - success = stripForward(currSeq); - if(!success){ trashCode += 'f'; } + success = stripForward(currSeq, currQual, groupPrime); + if(success > pdiffs) { trashCode += 'f'; } + else{ currentSeqsDiffs += success; } } - + + if (currentSeqsDiffs > tdiffs) { trashCode += 't'; } + if(numRPrimers != 0){ - success = stripReverse(currSeq); - if(!success){ trashCode += 'r'; } + success = stripReverse(currSeq, currQual); + if(!success) { trashCode += 'r'; } } if(minLength > 0 || maxLength > 0){ success = cullLength(currSeq); - if(!success){ trashCode += 'l'; } + if(!success) { trashCode += 'l'; } } if(maxHomoP > 0){ success = cullHomoP(currSeq); - if(!success){ trashCode += 'h'; } + if(!success) { trashCode += 'h'; } } if(maxAmbig != -1){ success = cullAmbigs(currSeq); - if(!success){ trashCode += 'n'; } + if(!success) { trashCode += 'n'; } } - if(flip){ currSeq.reverseComplement(); } // should go last + if(flip){ currSeq.reverseComplement(); currQual.flipQScores(); } // should go last if(trashCode.length() == 0){ - currSeq.setAligned(currSeq.getUnaligned()); //this is because of a modification we made to the sequence class to fix a bug. all seqs have an aligned version, which is the version that gets printed. + currSeq.setAligned(currSeq.getUnaligned()); currSeq.printSequence(outFASTA); + currQual.printQScores(outQual); + if(barcodes.size() != 0){ - outGroups << currSeq.getName() << '\t' << groupVector[group] << endl; - + string thisGroup = groupVector[groupBar]; + int indexToFastaFile = groupBar; + if (primers.size() != 0){ + //does this primer have a group + if (groupVector[groupPrime] != "") { + thisGroup += "." + groupVector[groupPrime]; + indexToFastaFile = combos[thisGroup]; + } + } + outGroups << currSeq.getName() << '\t' << thisGroup << endl; if(allFiles){ - currSeq.printSequence(*fastaFileNames[group]); + currSeq.printSequence(*fastaFileNames[indexToFastaFile]); + + if(qFileName != ""){ + currQual.printQScores(*qualFileNames[indexToFastaFile]); + } } } } else{ currSeq.setName(currSeq.getName() + '|' + trashCode); currSeq.setUnaligned(origSeq); + currSeq.setAligned(origSeq); currSeq.printSequence(scrapFASTA); + currQual.printQScores(scrapQual); } } gobble(inFASTA); + gobble(qFile); } inFASTA.close(); outFASTA.close(); scrapFASTA.close(); if (oligoFile != "") { outGroups.close(); } - if(qFileName != "") { qFile.close(); } + if(qFileName != "") { qFile.close(); scrapQual.close(); outQual.close(); } for(int i=0;iclose(); delete fastaFileNames[i]; } + if(qFileName != ""){ + for(int i=0;iclose(); + delete qualFileNames[i]; + } + } + return 0; } catch(exception& e) { @@ -430,8 +606,10 @@ int TrimSeqsCommand::driverCreateTrim(string filename, string qFileName, string exit(1); } } + /**************************************************************************************************/ -int TrimSeqsCommand::createProcessesCreateTrim(string filename, string qFileName, string trimFile, string scrapFile, string groupFile, vector fastaNames) { + +int TrimSeqsCommand::createProcessesCreateTrim(string filename, string qFileName, string trimFile, string scrapFile, string trimQFile, string scrapQFile, string groupFile, vector fastaNames, vector qualNames) { try { #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) int process = 0; @@ -446,7 +624,7 @@ int TrimSeqsCommand::createProcessesCreateTrim(string filename, string qFileName 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){ - driverCreateTrim(filename, qFileName, (trimFile + toString(getpid()) + ".temp"), (scrapFile + toString(getpid()) + ".temp"), (groupFile + toString(getpid()) + ".temp"), fastaNames, lines[process], qLines[process]); + driverCreateTrim(filename, qFileName, (trimFile + toString(getpid()) + ".temp"), (scrapFile + toString(getpid()) + ".temp"), (trimQFile + toString(getpid()) + ".temp"), (scrapQFile + toString(getpid()) + ".temp"), (groupFile + toString(getpid()) + ".temp"), fastaNames, qualNames, lines[process], qLines[process]); exit(0); }else { m->mothurOut("unable to spawn the necessary processes."); m->mothurOutEndLine(); exit(0); } } @@ -465,6 +643,7 @@ int TrimSeqsCommand::createProcessesCreateTrim(string filename, string qFileName exit(1); } } + /**************************************************************************************************/ int TrimSeqsCommand::setLines(string filename, vector& lines) { @@ -472,7 +651,7 @@ int TrimSeqsCommand::setLines(string filename, vector& lines) { lines.clear(); - vector positions; + vector positions; ifstream inFASTA; openInputFile(filename, inFASTA); @@ -482,7 +661,7 @@ int TrimSeqsCommand::setLines(string filename, vector& lines) { input = getline(inFASTA); if (input.length() != 0) { - if(input[0] == '>'){ long int pos = inFASTA.tellg(); positions.push_back(pos - input.length() - 1); } + if(input[0] == '>'){ unsigned long int pos = inFASTA.tellg(); positions.push_back(pos - input.length() - 1); } } } inFASTA.close(); @@ -490,7 +669,7 @@ int TrimSeqsCommand::setLines(string filename, vector& lines) { int numFastaSeqs = positions.size(); FILE * pFile; - long size; + unsigned long int size; //get num bytes in file pFile = fopen (filename.c_str(),"rb"); @@ -505,11 +684,11 @@ int TrimSeqsCommand::setLines(string filename, vector& lines) { for (int i = 0; i < processors; i++) { - long int startPos = positions[ i * numSeqsPerProcessor ]; + unsigned long int startPos = positions[ i * numSeqsPerProcessor ]; if(i == processors - 1){ numSeqsPerProcessor = numFastaSeqs - i * numSeqsPerProcessor; }else{ - long int myEnd = positions[ (i+1) * numSeqsPerProcessor ]; + unsigned long int myEnd = positions[ (i+1) * numSeqsPerProcessor ]; } lines.push_back(new linePair(startPos, numSeqsPerProcessor)); } @@ -523,7 +702,7 @@ int TrimSeqsCommand::setLines(string filename, vector& lines) { } //*************************************************************************************************************** -void TrimSeqsCommand::getOligos(vector& outFASTAVec){ //vector& outFASTAVec +void TrimSeqsCommand::getOligos(vector& outFASTAVec, vector& outQualVec){ try { ifstream inOligos; openInputFile(oligoFile, inOligos); @@ -532,14 +711,18 @@ void TrimSeqsCommand::getOligos(vector& outFASTAVec){ //vector> type; - + if(type[0] == '#'){ while (!inOligos.eof()) { char c = inOligos.get(); if (c == 10 || c == 13){ break; } } // get rest of line if there's any crap there } else{ + //make type case insensitive + for(int i=0;i> oligo; for(int i=0;i& outFASTAVec){ //vector::iterator itPrime = primers.find(oligo); + if (itPrime != primers.end()) { m->mothurOut("primer " + oligo + " is in your oligos file already."); m->mothurOutEndLine(); } + + primers[oligo]=index; index++; + groupVector.push_back(group); + + if(allFiles){ + outFASTAVec.push_back((outputDir + getRootName(getSimpleName(fastaFile)) + group + ".fasta")); + if(qFileName != ""){ + outQualVec.push_back((outputDir + getRootName(getSimpleName(qFileName)) + group + ".qual")); + } + if (group == "") { //if there is not a group for this primer, then this file will not get written to, but we add it to keep the indexes correct + filesToRemove.insert((outputDir + getRootName(getSimpleName(fastaFile)) + group + ".fasta")); + if(qFileName != ""){ + filesToRemove.insert((outputDir + getRootName(getSimpleName(qFileName)) + group + ".qual")); + } + }else { + outputNames.push_back((outputDir + getRootName(getSimpleName(fastaFile)) + group + ".fasta")); + if(qFileName != ""){ + outputNames.push_back((outputDir + getRootName(getSimpleName(qFileName)) + group + ".qual")); + } + } + } + } - else if(type == "reverse"){ + else if(type == "REVERSE"){ Sequence oligoRC("reverse", oligo); oligoRC.reverseComplement(); revPrimer.push_back(oligoRC.getUnaligned()); } - else if(type == "barcode"){ + else if(type == "BARCODE"){ inOligos >> group; - barcodes[oligo]=index++; + + //check for repeat barcodes + map::iterator itBar = barcodes.find(oligo); + if (itBar != barcodes.end()) { m->mothurOut("barcode " + oligo + " is in your oligos file already."); m->mothurOutEndLine(); } + + barcodes[oligo]=index; index++; groupVector.push_back(group); if(allFiles){ - //outFASTAVec.push_back(new ofstream((outputDir + getRootName(getSimpleName(fastaFile)) + group + ".fasta").c_str(), ios::ate)); outputNames.push_back((outputDir + getRootName(getSimpleName(fastaFile)) + group + ".fasta")); outFASTAVec.push_back((outputDir + getRootName(getSimpleName(fastaFile)) + group + ".fasta")); + if(qFileName != ""){ + outQualVec.push_back((outputDir + getRootName(getSimpleName(qFileName)) + group + ".qual")); + outputNames.push_back((outputDir + getRootName(getSimpleName(qFileName)) + group + ".qual")); + } } - } + }else{ m->mothurOut(type + " is not recognized as a valid type. Choices are forward, reverse, and barcode. Ignoring " + oligo + "."); m->mothurOutEndLine(); } } + gobble(inOligos); } inOligos.close(); - numFPrimers = forPrimer.size(); + //add in potential combos + if(allFiles){ + comboStarts = outFASTAVec.size()-1; + for (map::iterator itBar = barcodes.begin(); itBar != barcodes.end(); itBar++) { + for (map::iterator itPrime = primers.begin(); itPrime != primers.end(); itPrime++) { + if (groupVector[itPrime->second] != "") { //there is a group for this primer + outputNames.push_back((outputDir + getRootName(getSimpleName(qFileName)) + groupVector[itBar->second] + "." + groupVector[itPrime->second] + ".fasta")); + outFASTAVec.push_back((outputDir + getRootName(getSimpleName(fastaFile)) + groupVector[itBar->second] + "." + groupVector[itPrime->second] + ".fasta")); + combos[(groupVector[itBar->second] + "." + groupVector[itPrime->second])] = outFASTAVec.size()-1; + + if(qFileName != ""){ + outQualVec.push_back((outputDir + getRootName(getSimpleName(qFileName)) + groupVector[itBar->second] + "." + groupVector[itPrime->second] + ".qual")); + outputNames.push_back((outputDir + getRootName(getSimpleName(qFileName)) + groupVector[itBar->second] + "." + groupVector[itPrime->second] + ".qual")); + } + } + } + } + } + + numFPrimers = primers.size(); numRPrimers = revPrimer.size(); } @@ -579,30 +824,125 @@ void TrimSeqsCommand::getOligos(vector& outFASTAVec){ //vectorerrorOut(e, "TrimSeqsCommand", "getOligos"); exit(1); } - } - //*************************************************************************************************************** -bool TrimSeqsCommand::stripBarcode(Sequence& seq, int& group){ +int TrimSeqsCommand::stripBarcode(Sequence& seq, QualityScores& qual, int& group){ try { + string rawSequence = seq.getUnaligned(); - bool success = 0; //guilty until proven innocent + int success = bdiffs + 1; //guilty until proven innocent + //can you find the barcode for(map::iterator it=barcodes.begin();it!=barcodes.end();it++){ string oligo = it->first; if(rawSequence.length() < oligo.length()){ //let's just assume that the barcodes are the same length - success = 0; - break; + success = bdiffs + 10; //if the sequence is shorter than the barcode then bail out + break; } if(compareDNASeq(oligo, rawSequence.substr(0,oligo.length()))){ group = it->second; seq.setUnaligned(rawSequence.substr(oligo.length())); - success = 1; + + if(qual.getName() != ""){ + qual.trimQScores(oligo.length(), -1); + } + + success = 0; break; } } + + //if you found the barcode or if you don't want to allow for diffs +// cout << success; + if ((bdiffs == 0) || (success == 0)) { return success; } + + else { //try aligning and see if you can find it +// cout << endl; + + int maxLength = 0; + + Alignment* alignment; + if (barcodes.size() > 0) { + map::iterator it=barcodes.begin(); + + for(it;it!=barcodes.end();it++){ + if(it->first.length() > maxLength){ + maxLength = it->first.length(); + } + } + alignment = new NeedlemanOverlap(-1.0, 1.0, -1.0, (maxLength+bdiffs+1)); + + }else{ alignment = NULL; } + + //can you find the barcode + int minDiff = 1e6; + int minCount = 1; + int minGroup = -1; + int minPos = 0; + + for(map::iterator it=barcodes.begin();it!=barcodes.end();it++){ + string oligo = it->first; +// int length = oligo.length(); + + if(rawSequence.length() < maxLength){ //let's just assume that the barcodes are the same length + success = bdiffs + 10; + break; + } + + //use needleman to align first barcode.length()+numdiffs of sequence to each barcode + alignment->align(oligo, rawSequence.substr(0,oligo.length()+bdiffs)); + oligo = alignment->getSeqAAln(); + string temp = alignment->getSeqBAln(); + + int alnLength = oligo.length(); + + for(int i=oligo.length()-1;i>=0;i--){ + if(oligo[i] != '-'){ alnLength = i+1; break; } + } + oligo = oligo.substr(0,alnLength); + temp = temp.substr(0,alnLength); + + int newStart=0; + int numDiff = countDiffs(oligo, temp); + +// cout << oligo << '\t' << temp << '\t' << numDiff << endl; + + if(numDiff < minDiff){ + minDiff = numDiff; + minCount = 1; + minGroup = it->second; + minPos = 0; + for(int i=0;i bdiffs) { success = minDiff; } //no good matches + else if(minCount > 1) { success = bdiffs + 100; } //can't tell the difference between multiple barcodes + else{ //use the best match + group = minGroup; + seq.setUnaligned(rawSequence.substr(minPos)); + + if(qual.getName() != ""){ + qual.trimQScores(minPos, -1); + } + success = minDiff; + } + + if (alignment != NULL) { delete alignment; } + + } +// cout << success << endl; + return success; } @@ -615,28 +955,120 @@ bool TrimSeqsCommand::stripBarcode(Sequence& seq, int& group){ //*************************************************************************************************************** -bool TrimSeqsCommand::stripForward(Sequence& seq){ +int TrimSeqsCommand::stripForward(Sequence& seq, QualityScores& qual, int& group){ try { string rawSequence = seq.getUnaligned(); - bool success = 0; //guilty until proven innocent + int success = pdiffs + 1; //guilty until proven innocent - for(int i=0;i::iterator it=primers.begin();it!=primers.end();it++){ + string oligo = it->first; + if(rawSequence.length() < oligo.length()){ //let's just assume that the primers are the same length + success = pdiffs + 10; //if the sequence is shorter than the barcode then bail out + break; } if(compareDNASeq(oligo, rawSequence.substr(0,oligo.length()))){ + group = it->second; seq.setUnaligned(rawSequence.substr(oligo.length())); - success = 1; + if(qual.getName() != ""){ + qual.trimQScores(oligo.length(), -1); + + } + success = 0; break; } } + + //if you found the barcode or if you don't want to allow for diffs +// cout << success; + if ((pdiffs == 0) || (success == 0)) { return success; } - return success; + else { //try aligning and see if you can find it +// cout << endl; + + int maxLength = 0; + + Alignment* alignment; + if (primers.size() > 0) { + map::iterator it=primers.begin(); + + for(it;it!=primers.end();it++){ + if(it->first.length() > maxLength){ + maxLength = it->first.length(); + } + } + alignment = new NeedlemanOverlap(-1.0, 1.0, -1.0, (maxLength+pdiffs+1)); + + }else{ alignment = NULL; } + + //can you find the barcode + int minDiff = 1e6; + int minCount = 1; + int minGroup = -1; + int minPos = 0; + + for(map::iterator it=primers.begin();it!=primers.end();it++){ + string oligo = it->first; +// int length = oligo.length(); + + if(rawSequence.length() < maxLength){ + success = pdiffs + 100; + break; + } + + //use needleman to align first barcode.length()+numdiffs of sequence to each barcode + alignment->align(oligo, rawSequence.substr(0,oligo.length()+pdiffs)); + oligo = alignment->getSeqAAln(); + string temp = alignment->getSeqBAln(); + int alnLength = oligo.length(); + + for(int i=oligo.length()-1;i>=0;i--){ + if(oligo[i] != '-'){ alnLength = i+1; break; } + } + oligo = oligo.substr(0,alnLength); + temp = temp.substr(0,alnLength); + + int newStart=0; + int numDiff = countDiffs(oligo, temp); + +// cout << oligo << '\t' << temp << '\t' << numDiff << endl; + + if(numDiff < minDiff){ + minDiff = numDiff; + minCount = 1; + minGroup = it->second; + minPos = 0; + for(int i=0;i pdiffs) { success = minDiff; } //no good matches + else if(minCount > 1) { success = pdiffs + 10; } //can't tell the difference between multiple primers + else{ //use the best match + group = minGroup; + seq.setUnaligned(rawSequence.substr(minPos)); + if(qual.getName() != ""){ + qual.trimQScores(minPos, -1); + } + success = minDiff; + } + + if (alignment != NULL) { delete alignment; } + + } + + return success; + } catch(exception& e) { m->errorOut(e, "TrimSeqsCommand", "stripForward"); @@ -646,7 +1078,7 @@ bool TrimSeqsCommand::stripForward(Sequence& seq){ //*************************************************************************************************************** -bool TrimSeqsCommand::stripReverse(Sequence& seq){ +bool TrimSeqsCommand::stripReverse(Sequence& seq, QualityScores& qual){ try { string rawSequence = seq.getUnaligned(); bool success = 0; //guilty until proven innocent @@ -661,6 +1093,9 @@ bool TrimSeqsCommand::stripReverse(Sequence& seq){ if(compareDNASeq(oligo, rawSequence.substr(rawSequence.length()-oligo.length(),oligo.length()))){ seq.setUnaligned(rawSequence.substr(0,rawSequence.length()-oligo.length())); + if(qual.getName() != ""){ + qual.trimQScores(-1, rawSequence.length()-oligo.length()); + } success = 1; break; } @@ -744,7 +1179,7 @@ bool TrimSeqsCommand::compareDNASeq(string oligo, string seq){ for(int i=0;i> name; - if (name.length() != 0) { if(name.substr(1) != seq.getName()) { m->mothurOut("sequence name mismatch btwn fasta and qual file"); m->mothurOutEndLine(); } } - while (!qFile.eof()) { char c = qFile.get(); if (c == 10 || c == 13){ break; } } - - int score; - int end = seqLength; + + int length = oligo.length(); + int countDiffs = 0; - for(int i=0;i> score; - - if(score <= qThreshold){ - end = i; - break; + for(int i=0;i> score; - } - - seq.setUnaligned(rawSequence.substr(0,end)); - return 1; + return countDiffs; } catch(exception& e) { - m->errorOut(e, "TrimSeqsCommand", "stripQualThreshold"); + m->errorOut(e, "TrimSeqsCommand", "countDiffs"); exit(1); } -} +} //*************************************************************************************************************** -bool TrimSeqsCommand::cullQualAverage(Sequence& seq, ifstream& qFile){ - try { - string rawSequence = seq.getUnaligned(); - int seqLength = seq.getNumBases(); - bool success = 0; //guilty until proven innocent - string name; - - qFile >> name; - if (name[0] == '>') { if(name.substr(1) != seq.getName()) { m->mothurOut("sequence name mismatch btwn fasta: " + seq.getName() + " and qual file: " + name); m->mothurOutEndLine(); } } - - while (!qFile.eof()) { char c = qFile.get(); if (c == 10 || c == 13){ break; } } - - float score; - float average = 0; - - for(int i=0;i> score; - average += score; - } - average /= seqLength; +//bool TrimSeqsCommand::stripQualThreshold(Sequence& seq, ifstream& qFile){ +// try { +// +// string rawSequence = seq.getUnaligned(); +// int seqLength = seq.getNumBases(); +// bool success = 0; //guilty until proven innocent +// string name; +// +// qFile >> name; +// if (name[0] == '>') { if(name.substr(1) != seq.getName()) { m->mothurOut("sequence name mismatch btwn fasta: " + seq.getName() + " and qual file: " + name); m->mothurOutEndLine(); } } +// +// while (!qFile.eof()) { char c = qFile.get(); if (c == 10 || c == 13){ break; } } +// +// int score; +// int end = seqLength; +// +// for(int i=0;i> score; +// +// if(score < qThreshold){ +// end = i; +// break; +// } +// } +// for(int i=end+1;i> score; +// } +// +// seq.setUnaligned(rawSequence.substr(0,end)); +// +// return 1; +// } +// catch(exception& e) { +// m->errorOut(e, "TrimSeqsCommand", "stripQualThreshold"); +// exit(1); +// } +//} - if(average >= qAverage) { success = 1; } - else { success = 0; } - - return success; - } - catch(exception& e) { - m->errorOut(e, "TrimSeqsCommand", "cullQualAverage"); - exit(1); - } -} +//*************************************************************************************************************** + +//bool TrimSeqsCommand::cullQualAverage(Sequence& seq, ifstream& qFile){ +// try { +// string rawSequence = seq.getUnaligned(); +// int seqLength = seq.getNumBases(); +// bool success = 0; //guilty until proven innocent +// string name; +// +// qFile >> name; +// if (name[0] == '>') { if(name.substr(1) != seq.getName()) { m->mothurOut("sequence name mismatch btwn fasta: " + seq.getName() + " and qual file: " + name); m->mothurOutEndLine(); } } +// +// while (!qFile.eof()) { char c = qFile.get(); if (c == 10 || c == 13){ break; } } +// +// float score; +// float average = 0; +// +// for(int i=0;i> score; +// average += score; +// } +// average /= seqLength; +// +// if(average >= qAverage) { success = 1; } +// else { success = 0; } +// +// return success; +// } +// catch(exception& e) { +// m->errorOut(e, "TrimSeqsCommand", "cullQualAverage"); +// exit(1); +// } +//} //***************************************************************************************************************