X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=trimseqscommand.cpp;h=6a8afd017c1e88ff1bdbde883608b5eff00b235f;hb=49d2b7459c5027557564b21e9487dadafbbbdc96;hp=ed84cc0c60d404b0d7c12eb0c3751a2e2a601706;hpb=62c36830aae6dd6151898ec6e07df59c8aed79fe;p=mothur.git diff --git a/trimseqscommand.cpp b/trimseqscommand.cpp index ed84cc0..6a8afd0 100644 --- a/trimseqscommand.cpp +++ b/trimseqscommand.cpp @@ -9,6 +9,7 @@ #include "trimseqscommand.h" #include "needlemanoverlap.hpp" +#include "trimoligos.h" //********************************************************************************************************************** vector TrimSeqsCommand::setParameters(){ @@ -24,9 +25,12 @@ vector TrimSeqsCommand::setParameters(){ CommandParameter pmaxlength("maxlength", "Number", "", "0", "", "", "",false,false); parameters.push_back(pmaxlength); CommandParameter ppdiffs("pdiffs", "Number", "", "0", "", "", "",false,false); parameters.push_back(ppdiffs); CommandParameter pbdiffs("bdiffs", "Number", "", "0", "", "", "",false,false); parameters.push_back(pbdiffs); - CommandParameter ptdiffs("tdiffs", "Number", "", "0", "", "", "",false,false); parameters.push_back(ptdiffs); + CommandParameter pldiffs("ldiffs", "Number", "", "0", "", "", "",false,false); parameters.push_back(pldiffs); + CommandParameter psdiffs("sdiffs", "Number", "", "0", "", "", "",false,false); parameters.push_back(psdiffs); + CommandParameter ptdiffs("tdiffs", "Number", "", "0", "", "", "",false,false); parameters.push_back(ptdiffs); CommandParameter pprocessors("processors", "Number", "", "1", "", "", "",false,false); parameters.push_back(pprocessors); CommandParameter pallfiles("allfiles", "Boolean", "", "F", "", "", "",false,false); parameters.push_back(pallfiles); + CommandParameter pkeepforward("keepforward", "Boolean", "", "F", "", "", "",false,false); parameters.push_back(pkeepforward); CommandParameter pqtrim("qtrim", "Boolean", "", "T", "", "", "",false,false); parameters.push_back(pqtrim); CommandParameter pqthreshold("qthreshold", "Number", "", "0", "", "", "",false,false); parameters.push_back(pqthreshold); CommandParameter pqaverage("qaverage", "Number", "", "0", "", "", "",false,false); parameters.push_back(pqaverage); @@ -63,9 +67,11 @@ string TrimSeqsCommand::getHelpString(){ helpString += "The maxhomop parameter allows you to set a maximum homopolymer length. \n"; helpString += "The minlength parameter allows you to set and minimum sequence length. \n"; helpString += "The maxlength parameter allows you to set and maximum sequence length. \n"; - helpString += "The tdiffs parameter is used to specify the total number of differences allowed in the sequence. The default is pdiffs + bdiffs.\n"; + helpString += "The tdiffs parameter is used to specify the total number of differences allowed in the sequence. The default is pdiffs + bdiffs + sdiffs + ldiffs.\n"; helpString += "The bdiffs parameter is used to specify the number of differences allowed in the barcode. The default is 0.\n"; helpString += "The pdiffs parameter is used to specify the number of differences allowed in the primer. The default is 0.\n"; + helpString += "The ldiffs parameter is used to specify the number of differences allowed in the linker. The default is 0.\n"; + helpString += "The sdiffs parameter is used to specify the number of differences allowed in the spacer. The default is 0.\n"; helpString += "The qfile parameter allows you to provide a quality file.\n"; helpString += "The qthreshold parameter allows you to set a minimum quality score allowed. \n"; helpString += "The qaverage parameter allows you to set a minimum average quality score allowed. \n"; @@ -74,6 +80,7 @@ string TrimSeqsCommand::getHelpString(){ helpString += "The rollaverage parameter allows you to set a minimum rolling average quality score allowed over a window. \n"; helpString += "The qstepsize parameter allows you to set a number of bases to move the window over. Default=1.\n"; helpString += "The allfiles parameter will create separate group and fasta file for each grouping. The default is F.\n"; + helpString += "The keepforward parameter allows you to indicate whether you want the forward primer removed or not. The default is F, meaning remove the forward primer.\n"; helpString += "The qtrim parameter will trim sequence from the point that they fall below the qthreshold and put it in the .trim file if set to true. The default is T.\n"; helpString += "The keepfirst parameter trims the sequence to the first keepfirst number of bases after the barcode or primers are removed, before the sequence is checked to see if it meets the other requirements. \n"; helpString += "The removelast removes the last removelast number of bases after the barcode or primers are removed, before the sequence is checked to see if it meets the other requirements.\n"; @@ -90,6 +97,29 @@ string TrimSeqsCommand::getHelpString(){ exit(1); } } +//********************************************************************************************************************** +string TrimSeqsCommand::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 == "qfile") { outputFileName = "qual"; } + else if (type == "fasta") { outputFileName = "fasta"; } + else if (type == "group") { outputFileName = "groups"; } + else if (type == "name") { outputFileName = "names"; } + 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, "TrimSeqsCommand", "getOutputFileNameTag"); + exit(1); + } +} //********************************************************************************************************************** @@ -189,6 +219,7 @@ TrimSeqsCommand::TrimSeqsCommand(string option) { 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 if (fastaFile == "not open") { abort = true; } + else { m->setFastaFile(fastaFile); } //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"){ @@ -201,50 +232,56 @@ TrimSeqsCommand::TrimSeqsCommand(string option) { // ...at some point should added some additional type checking... string temp; temp = validParameter.validFile(parameters, "flip", false); - if (temp == "not found"){ flip = 0; } - else if(m->isTrue(temp)) { flip = 1; } + if (temp == "not found") { flip = 0; } + else { flip = m->isTrue(temp); } temp = validParameter.validFile(parameters, "oligos", true); if (temp == "not found"){ oligoFile = ""; } else if(temp == "not open"){ abort = true; } - else { oligoFile = temp; } + else { oligoFile = temp; m->setOligosFile(oligoFile); } temp = validParameter.validFile(parameters, "maxambig", false); if (temp == "not found") { temp = "-1"; } - convert(temp, maxAmbig); + m->mothurConvert(temp, maxAmbig); temp = validParameter.validFile(parameters, "maxhomop", false); if (temp == "not found") { temp = "0"; } - convert(temp, maxHomoP); + m->mothurConvert(temp, maxHomoP); temp = validParameter.validFile(parameters, "minlength", false); if (temp == "not found") { temp = "0"; } - convert(temp, minLength); + m->mothurConvert(temp, minLength); temp = validParameter.validFile(parameters, "maxlength", false); if (temp == "not found") { temp = "0"; } - convert(temp, maxLength); + m->mothurConvert(temp, maxLength); temp = validParameter.validFile(parameters, "bdiffs", false); if (temp == "not found") { temp = "0"; } - convert(temp, bdiffs); + m->mothurConvert(temp, bdiffs); temp = validParameter.validFile(parameters, "pdiffs", false); if (temp == "not found") { temp = "0"; } - convert(temp, pdiffs); + m->mothurConvert(temp, pdiffs); + + temp = validParameter.validFile(parameters, "ldiffs", false); if (temp == "not found") { temp = "0"; } + m->mothurConvert(temp, ldiffs); + + temp = validParameter.validFile(parameters, "sdiffs", false); if (temp == "not found") { temp = "0"; } + m->mothurConvert(temp, sdiffs); - temp = validParameter.validFile(parameters, "tdiffs", false); if (temp == "not found") { int tempTotal = pdiffs + bdiffs; temp = toString(tempTotal); } - convert(temp, tdiffs); + temp = validParameter.validFile(parameters, "tdiffs", false); if (temp == "not found") { int tempTotal = pdiffs + bdiffs + ldiffs + sdiffs; temp = toString(tempTotal); } + m->mothurConvert(temp, tdiffs); - if(tdiffs == 0){ tdiffs = bdiffs + pdiffs; } + if(tdiffs == 0){ tdiffs = bdiffs + pdiffs + ldiffs + sdiffs; } temp = validParameter.validFile(parameters, "qfile", true); if (temp == "not found") { qFileName = ""; } else if(temp == "not open") { abort = true; } - else { qFileName = temp; } + else { qFileName = temp; m->setQualFile(qFileName); } temp = validParameter.validFile(parameters, "name", true); if (temp == "not found") { nameFile = ""; } - else if(temp == "not open") { abort = true; } - else { nameFile = temp; } + else if(temp == "not open") { nameFile = ""; abort = true; } + else { nameFile = temp; m->setNameFile(nameFile); } temp = validParameter.validFile(parameters, "qthreshold", false); if (temp == "not found") { temp = "0"; } - convert(temp, qThreshold); + m->mothurConvert(temp, qThreshold); temp = validParameter.validFile(parameters, "qtrim", false); if (temp == "not found") { temp = "t"; } qtrim = m->isTrue(temp); @@ -272,10 +309,13 @@ TrimSeqsCommand::TrimSeqsCommand(string option) { temp = validParameter.validFile(parameters, "allfiles", false); if (temp == "not found") { temp = "F"; } allFiles = m->isTrue(temp); + + temp = validParameter.validFile(parameters, "keepforward", false); if (temp == "not found") { temp = "F"; } + keepforward = m->isTrue(temp); temp = validParameter.validFile(parameters, "processors", false); if (temp == "not found"){ temp = m->getProcessors(); } m->setProcessors(temp); - convert(temp, processors); + m->mothurConvert(temp, processors); if(allFiles && (oligoFile == "")){ @@ -290,6 +330,11 @@ TrimSeqsCommand::TrimSeqsCommand(string option) { m->mothurOut("You didn't set any options... quiting command."); m->mothurOutEndLine(); abort = true; } + + if (nameFile == "") { + vector files; files.push_back(fastaFile); + parser.getNameFile(files); + } } } @@ -307,18 +352,21 @@ int TrimSeqsCommand::execute(){ numFPrimers = 0; //this needs to be initialized numRPrimers = 0; + numSpacers = 0; + numLinkers = 0; + createGroup = false; vector > fastaFileNames; vector > qualFileNames; vector > nameFileNames; - string trimSeqFile = outputDir + m->getRootName(m->getSimpleName(fastaFile)) + "trim.fasta"; + string trimSeqFile = outputDir + m->getRootName(m->getSimpleName(fastaFile)) + "trim." + getOutputFileNameTag("fasta"); outputNames.push_back(trimSeqFile); outputTypes["fasta"].push_back(trimSeqFile); - string scrapSeqFile = outputDir + m->getRootName(m->getSimpleName(fastaFile)) + "scrap.fasta"; + string scrapSeqFile = outputDir + m->getRootName(m->getSimpleName(fastaFile)) + "scrap." + getOutputFileNameTag("fasta"); outputNames.push_back(scrapSeqFile); outputTypes["fasta"].push_back(scrapSeqFile); - string trimQualFile = outputDir + m->getRootName(m->getSimpleName(fastaFile)) + "trim.qual"; - string scrapQualFile = outputDir + m->getRootName(m->getSimpleName(fastaFile)) + "scrap.qual"; + string trimQualFile = outputDir + m->getRootName(m->getSimpleName(fastaFile)) + "trim." + getOutputFileNameTag("qfile"); + string scrapQualFile = outputDir + m->getRootName(m->getSimpleName(fastaFile)) + "scrap." + getOutputFileNameTag("qfile"); if (qFileName != "") { outputNames.push_back(trimQualFile); @@ -327,8 +375,8 @@ int TrimSeqsCommand::execute(){ outputTypes["qfile"].push_back(scrapQualFile); } - string trimNameFile = outputDir + m->getRootName(m->getSimpleName(nameFile)) + "trim.names"; - string scrapNameFile = outputDir + m->getRootName(m->getSimpleName(nameFile)) + "scrap.names"; + string trimNameFile = outputDir + m->getRootName(m->getSimpleName(nameFile)) + "trim." + getOutputFileNameTag("name"); + string scrapNameFile = outputDir + m->getRootName(m->getSimpleName(nameFile)) + "scrap." + getOutputFileNameTag("name"); if (nameFile != "") { m->readNames(nameFile, nameMap); @@ -342,31 +390,22 @@ int TrimSeqsCommand::execute(){ string outputGroupFileName; if(oligoFile != ""){ - outputGroupFileName = outputDir + m->getRootName(m->getSimpleName(fastaFile)) + "groups"; - outputNames.push_back(outputGroupFileName); outputTypes["group"].push_back(outputGroupFileName); - getOligos(fastaFileNames, qualFileNames, nameFileNames); + createGroup = getOligos(fastaFileNames, qualFileNames, nameFileNames); + if (createGroup) { + outputGroupFileName = outputDir + m->getRootName(m->getSimpleName(fastaFile)) + getOutputFileNameTag("group"); + outputNames.push_back(outputGroupFileName); outputTypes["group"].push_back(outputGroupFileName); + } } + + //fills lines and qlines + setLines(fastaFile, qFileName); - vector fastaFilePos; - vector qFilePos; + if(processors == 1){ + driverCreateTrim(fastaFile, qFileName, trimSeqFile, scrapSeqFile, trimQualFile, scrapQualFile, trimNameFile, scrapNameFile, outputGroupFileName, fastaFileNames, qualFileNames, nameFileNames, lines[0], qLines[0]); + }else{ + createProcessesCreateTrim(fastaFile, qFileName, trimSeqFile, scrapSeqFile, trimQualFile, scrapQualFile, trimNameFile, scrapNameFile, outputGroupFileName, fastaFileNames, qualFileNames, nameFileNames); + } - setLines(fastaFile, qFileName, fastaFilePos, qFilePos); - - for (int i = 0; i < (fastaFilePos.size()-1); i++) { - lines.push_back(new linePair(fastaFilePos[i], fastaFilePos[(i+1)])); - if (qFileName != "") { qLines.push_back(new linePair(qFilePos[i], qFilePos[(i+1)])); } - } - if(qFileName == "") { qLines = lines; } //files with duds - - #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) - if(processors == 1){ - driverCreateTrim(fastaFile, qFileName, trimSeqFile, scrapSeqFile, trimQualFile, scrapQualFile, trimNameFile, scrapNameFile, outputGroupFileName, fastaFileNames, qualFileNames, nameFileNames, lines[0], qLines[0]); - }else{ - createProcessesCreateTrim(fastaFile, qFileName, trimSeqFile, scrapSeqFile, trimQualFile, scrapQualFile, trimNameFile, scrapNameFile, outputGroupFileName, fastaFileNames, qualFileNames, nameFileNames); - } - #else - driverCreateTrim(fastaFile, qFileName, trimSeqFile, scrapSeqFile, trimQualFile, scrapQualFile, trimNameFile, scrapNameFile, outputGroupFileName, fastaFileNames, qualFileNames, nameFileNames, lines[0], qLines[0]); - #endif if (m->control_pressed) { return 0; } @@ -379,16 +418,16 @@ int TrimSeqsCommand::execute(){ if (fastaFileNames[i][j] != "") { if (namesToRemove.count(fastaFileNames[i][j]) == 0) { if(m->isBlank(fastaFileNames[i][j])){ - remove(fastaFileNames[i][j].c_str()); + m->mothurRemove(fastaFileNames[i][j]); namesToRemove.insert(fastaFileNames[i][j]); if(qFileName != ""){ - remove(qualFileNames[i][j].c_str()); + m->mothurRemove(qualFileNames[i][j]); namesToRemove.insert(qualFileNames[i][j]); } if(nameFile != ""){ - remove(nameFileNames[i][j].c_str()); + m->mothurRemove(nameFileNames[i][j]); namesToRemove.insert(nameFileNames[i][j]); } }else{ @@ -412,7 +451,7 @@ int TrimSeqsCommand::execute(){ m->openInputFile(it->first, in); ofstream out; - string thisGroupName = outputDir + m->getRootName(m->getSimpleName(it->first)) + "groups"; + string thisGroupName = outputDir + m->getRootName(m->getSimpleName(it->first)) + getOutputFileNameTag("group"); outputNames.push_back(thisGroupName); outputTypes["group"].push_back(thisGroupName); m->openOutputFile(thisGroupName, out); @@ -421,23 +460,35 @@ int TrimSeqsCommand::execute(){ Sequence currSeq(in); m->gobble(in); out << currSeq.getName() << '\t' << it->second << endl; + + if (nameFile != "") { + map::iterator itName = nameMap.find(currSeq.getName()); + if (itName != nameMap.end()) { + vector thisSeqsNames; + m->splitAtChar(itName->second, thisSeqsNames, ','); + for (int k = 1; k < thisSeqsNames.size(); k++) { //start at 1 to skip self + out << thisSeqsNames[k] << '\t' << it->second << endl; + } + }else { m->mothurOut("[ERROR]: " + currSeq.getName() + " is not in your namefile, please correct."); m->mothurOutEndLine(); } + } } in.close(); out.close(); } } - if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } return 0; } + if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } return 0; } //output group counts m->mothurOutEndLine(); int total = 0; + if (groupCounts.size() != 0) { m->mothurOut("Group count: \n"); } for (map::iterator it = groupCounts.begin(); it != groupCounts.end(); it++) { - total += it->second; m->mothurOut("Group " + it->first + " contains " + toString(it->second) + " sequences."); m->mothurOutEndLine(); + total += it->second; m->mothurOut(it->first + "\t" + toString(it->second)); m->mothurOutEndLine(); } if (total != 0) { m->mothurOut("Total of all groups is " + toString(total)); m->mothurOutEndLine(); } - if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } return 0; } + if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } return 0; } //set fasta file as new current fastafile string current = ""; @@ -477,7 +528,7 @@ int TrimSeqsCommand::execute(){ /**************************************************************************************/ -int TrimSeqsCommand::driverCreateTrim(string filename, string qFileName, string trimFileName, string scrapFileName, string trimQFileName, string scrapQFileName, string trimNFileName, string scrapNFileName, string groupFileName, vector > fastaFileNames, vector > qualFileNames, vector > nameFileNames, linePair* line, linePair* qline) { +int TrimSeqsCommand::driverCreateTrim(string filename, string qFileName, string trimFileName, string scrapFileName, string trimQFileName, string scrapQFileName, string trimNFileName, string scrapNFileName, string groupFileName, vector > fastaFileNames, vector > qualFileNames, vector > nameFileNames, linePair line, linePair qline) { try { @@ -503,7 +554,7 @@ int TrimSeqsCommand::driverCreateTrim(string filename, string qFileName, string ofstream outGroupsFile; - if (oligoFile != ""){ m->openOutputFile(groupFileName, outGroupsFile); } + if (createGroup){ m->openOutputFile(groupFileName, outGroupsFile); } if(allFiles){ for (int i = 0; i < fastaFileNames.size(); i++) { //clears old file for (int j = 0; j < fastaFileNames[i].size(); j++) { //clears old file @@ -524,27 +575,28 @@ int TrimSeqsCommand::driverCreateTrim(string filename, string qFileName, string ifstream inFASTA; m->openInputFile(filename, inFASTA); - inFASTA.seekg(line->start); + inFASTA.seekg(line.start); ifstream qFile; if(qFileName != "") { m->openInputFile(qFileName, qFile); - qFile.seekg(qline->start); + qFile.seekg(qline.start); } int count = 0; bool moreSeqs = 1; + TrimOligos trimOligos(pdiffs, bdiffs, ldiffs, sdiffs, primers, barcodes, rbarcodes, revPrimer, linker, spacer); while (moreSeqs) { if (m->control_pressed) { inFASTA.close(); trimFASTAFile.close(); scrapFASTAFile.close(); - if (oligoFile != "") { outGroupsFile.close(); } + if (createGroup) { outGroupsFile.close(); } if(qFileName != ""){ qFile.close(); } - for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } + for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } return 0; } @@ -554,26 +606,46 @@ int TrimSeqsCommand::driverCreateTrim(string filename, string qFileName, string int currentSeqsDiffs = 0; Sequence currSeq(inFASTA); m->gobble(inFASTA); - + //cout << currSeq.getName() << '\t' << currSeq.getUnaligned().length() << endl; QualityScores currQual; if(qFileName != ""){ currQual = QualityScores(qFile); m->gobble(qFile); } - + string origSeq = currSeq.getUnaligned(); if (origSeq != "") { int barcodeIndex = 0; int primerIndex = 0; + if(numLinkers != 0){ + success = trimOligos.stripLinker(currSeq, currQual); + if(success > ldiffs) { trashCode += 'k'; } + else{ currentSeqsDiffs += success; } + + } + if(barcodes.size() != 0){ - success = stripBarcode(currSeq, currQual, barcodeIndex); + success = trimOligos.stripBarcode(currSeq, currQual, barcodeIndex); + if(success > bdiffs) { trashCode += 'b'; } + else{ currentSeqsDiffs += success; } + } + + if(rbarcodes.size() != 0){ + success = trimOligos.stripRBarcode(currSeq, currQual, barcodeIndex); if(success > bdiffs) { trashCode += 'b'; } else{ currentSeqsDiffs += success; } } + if(numSpacers != 0){ + success = trimOligos.stripSpacer(currSeq, currQual); + if(success > sdiffs) { trashCode += 's'; } + else{ currentSeqsDiffs += success; } + + } + if(numFPrimers != 0){ - success = stripForward(currSeq, currQual, primerIndex); + success = trimOligos.stripForward(currSeq, currQual, primerIndex, keepforward); if(success > pdiffs) { trashCode += 'f'; } else{ currentSeqsDiffs += success; } } @@ -581,7 +653,7 @@ int TrimSeqsCommand::driverCreateTrim(string filename, string qFileName, string if (currentSeqsDiffs > tdiffs) { trashCode += 't'; } if(numRPrimers != 0){ - success = stripReverse(currSeq, currQual); + success = trimOligos.stripReverse(currSeq, currQual); if(!success) { trashCode += 'r'; } } @@ -603,7 +675,7 @@ int TrimSeqsCommand::driverCreateTrim(string filename, string qFileName, string else if(qRollAverage != 0) { success = currQual.stripQualRollingAverage(currSeq, qRollAverage); } else if(qWindowAverage != 0){ success = currQual.stripQualWindowAverage(currSeq, qWindowStep, qWindowSize, qWindowAverage); } else { success = 1; } - + //you don't want to trim, if it fails above then scrap it if ((!qtrim) && (origLength != currSeq.getNumBases())) { success = 0; } @@ -638,25 +710,48 @@ int TrimSeqsCommand::driverCreateTrim(string filename, string qFileName, string currQual.printQScores(trimQualFile); } + if(nameFile != ""){ map::iterator itName = nameMap.find(currSeq.getName()); if (itName != nameMap.end()) { trimNameFile << itName->first << '\t' << itName->second << endl; } else { m->mothurOut("[ERROR]: " + currSeq.getName() + " is not in your namefile, please correct."); m->mothurOutEndLine(); } } - if(barcodes.size() != 0){ - string thisGroup = barcodeNameVector[barcodeIndex]; - if (primers.size() != 0) { if (primerNameVector[primerIndex] != "") { thisGroup += "." + primerNameVector[primerIndex]; } } - - outGroupsFile << currSeq.getName() << '\t' << thisGroup << endl; - - map::iterator it = groupCounts.find(thisGroup); - if (it == groupCounts.end()) { groupCounts[thisGroup] = 1; } - else { groupCounts[it->first]++; } + if (createGroup) { + if(barcodes.size() != 0){ + string thisGroup = barcodeNameVector[barcodeIndex]; + if (primers.size() != 0) { + if (primerNameVector[primerIndex] != "") { + if(thisGroup != "") { + thisGroup += "." + primerNameVector[primerIndex]; + }else { + thisGroup = primerNameVector[primerIndex]; + } + } + } + + outGroupsFile << currSeq.getName() << '\t' << thisGroup << endl; + int numRedundants = 0; + if (nameFile != "") { + map::iterator itName = nameMap.find(currSeq.getName()); + if (itName != nameMap.end()) { + vector thisSeqsNames; + m->splitAtChar(itName->second, thisSeqsNames, ','); + numRedundants = thisSeqsNames.size()-1; //we already include ourselves below + for (int k = 1; k < thisSeqsNames.size(); k++) { //start at 1 to skip self + outGroupsFile << thisSeqsNames[k] << '\t' << thisGroup << endl; + } + }else { m->mothurOut("[ERROR]: " + currSeq.getName() + " is not in your namefile, please correct."); m->mothurOutEndLine(); } + } + + map::iterator it = groupCounts.find(thisGroup); + if (it == groupCounts.end()) { groupCounts[thisGroup] = 1 + numRedundants; } + else { groupCounts[it->first] += (1 + numRedundants); } + + } } - if(allFiles){ ofstream output; m->openOutputFileAppend(fastaFileNames[barcodeIndex][primerIndex], output); @@ -696,25 +791,26 @@ int TrimSeqsCommand::driverCreateTrim(string filename, string qFileName, string count++; } - #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) - unsigned long int pos = inFASTA.tellg(); - if ((pos == -1) || (pos >= line->end)) { break; } + #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) || (__linux__) || (__unix__) || (__unix) + unsigned long long pos = inFASTA.tellg(); + if ((pos == -1) || (pos >= line.end)) { break; } + #else if (inFASTA.eof()) { break; } #endif - + //report progress if((count) % 1000 == 0){ m->mothurOut(toString(count)); m->mothurOutEndLine(); } } //report progress if((count) % 1000 != 0){ m->mothurOut(toString(count)); m->mothurOutEndLine(); } - + inFASTA.close(); trimFASTAFile.close(); scrapFASTAFile.close(); - if (oligoFile != "") { outGroupsFile.close(); } + if (createGroup) { outGroupsFile.close(); } if(qFileName != "") { qFile.close(); scrapQualFile.close(); trimQualFile.close(); } if(nameFile != "") { scrapNameFile.close(); trimNameFile.close(); } @@ -730,12 +826,13 @@ int TrimSeqsCommand::driverCreateTrim(string filename, string qFileName, string int TrimSeqsCommand::createProcessesCreateTrim(string filename, string qFileName, string trimFASTAFileName, string scrapFASTAFileName, string trimQualFileName, string scrapQualFileName, string trimNameFileName, string scrapNameFileName, string groupFile, vector > fastaFileNames, vector > qualFileNames, vector > nameFileNames) { try { -#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) - int process = 1; + + int process = 1; int exitCommand = 1; processIDS.clear(); - //loop through and create all the processes you want +#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) || (__linux__) || (__unix__) || (__unix) + //loop through and create all the processes you want while (process != processors) { int pid = fork(); @@ -786,14 +883,18 @@ int TrimSeqsCommand::createProcessesCreateTrim(string filename, string qFileName qLines[process]); //pass groupCounts to parent - ofstream out; - string tempFile = filename + toString(getpid()) + ".num.temp"; - m->openOutputFile(tempFile, out); - for (map::iterator it = groupCounts.begin(); it != groupCounts.end(); it++) { - out << it->first << '\t' << it->second << endl; + if(createGroup){ + ofstream out; + string tempFile = filename + toString(getpid()) + ".num.temp"; + m->openOutputFile(tempFile, out); + + out << groupCounts.size() << endl; + + for (map::iterator it = groupCounts.begin(); it != groupCounts.end(); it++) { + out << it->first << '\t' << it->second << endl; + } + out.close(); } - out.close(); - exit(0); }else { m->mothurOut("[ERROR]: unable to spawn the necessary processes."); m->mothurOutEndLine(); @@ -806,10 +907,14 @@ int TrimSeqsCommand::createProcessesCreateTrim(string filename, string qFileName ofstream temp; m->openOutputFile(trimFASTAFileName, temp); temp.close(); m->openOutputFile(scrapFASTAFileName, temp); temp.close(); - m->openOutputFile(trimQualFileName, temp); temp.close(); - m->openOutputFile(scrapQualFileName, temp); temp.close(); - m->openOutputFile(trimNameFileName, temp); temp.close(); - m->openOutputFile(scrapNameFileName, temp); temp.close(); + if(qFileName != ""){ + m->openOutputFile(trimQualFileName, temp); temp.close(); + m->openOutputFile(scrapQualFileName, temp); temp.close(); + } + if (nameFile != "") { + m->openOutputFile(trimNameFileName, temp); temp.close(); + m->openOutputFile(scrapNameFileName, temp); temp.close(); + } driverCreateTrim(filename, qFileName, trimFASTAFileName, scrapFASTAFileName, trimQualFileName, scrapQualFileName, trimNameFileName, scrapNameFileName, groupFile, fastaFileNames, qualFileNames, nameFileNames, lines[0], qLines[0]); @@ -818,33 +923,132 @@ int TrimSeqsCommand::createProcessesCreateTrim(string filename, string qFileName int temp = processIDS[i]; wait(&temp); } - - //append files +#else + ////////////////////////////////////////////////////////////////////////////////////////////////////// + //Windows version shared memory, so be careful when passing variables through the trimData struct. + //Above fork() will clone, so memory is separate, but that's not the case with windows, + ////////////////////////////////////////////////////////////////////////////////////////////////////// + + vector pDataArray; + DWORD dwThreadIdArray[processors-1]; + HANDLE hThreadArray[processors-1]; + + //Create processor worker threads. + for( int i=0; i > tempFASTAFileNames = fastaFileNames; + vector > tempPrimerQualFileNames = qualFileNames; + vector > tempNameFileNames = nameFileNames; + + if(allFiles){ + ofstream temp; + + for(int i=0;iopenOutputFile(tempFASTAFileNames[i][j], temp); temp.close(); + + if(qFileName != ""){ + tempPrimerQualFileNames[i][j] += extension; + m->openOutputFile(tempPrimerQualFileNames[i][j], temp); temp.close(); + } + if(nameFile != ""){ + tempNameFileNames[i][j] += extension; + m->openOutputFile(tempNameFileNames[i][j], temp); temp.close(); + } + } + } + } + } + + + trimData* tempTrim = new trimData(filename, + qFileName, nameFile, + (trimFASTAFileName+extension), + (scrapFASTAFileName+extension), + (trimQualFileName+extension), + (scrapQualFileName+extension), + (trimNameFileName+extension), + (scrapNameFileName+extension), + (groupFile+extension), + tempFASTAFileNames, + tempPrimerQualFileNames, + tempNameFileNames, + lines[i].start, lines[i].end, qLines[i].start, qLines[i].end, m, + pdiffs, bdiffs, ldiffs, sdiffs, tdiffs, primers, barcodes, rbarcodes, revPrimer, linker, spacer, + primerNameVector, barcodeNameVector, createGroup, allFiles, keepforward, keepFirst, removeLast, + qWindowStep, qWindowSize, qWindowAverage, qtrim, qThreshold, qAverage, qRollAverage, + minLength, maxAmbig, maxHomoP, maxLength, flip, nameMap); + pDataArray.push_back(tempTrim); + + hThreadArray[i] = CreateThread(NULL, 0, MyTrimThreadFunction, pDataArray[i], 0, &dwThreadIdArray[i]); + } + + //parent do my part + ofstream temp; + m->openOutputFile(trimFASTAFileName, temp); temp.close(); + m->openOutputFile(scrapFASTAFileName, temp); temp.close(); + if(qFileName != ""){ + m->openOutputFile(trimQualFileName, temp); temp.close(); + m->openOutputFile(scrapQualFileName, temp); temp.close(); + } + if (nameFile != "") { + m->openOutputFile(trimNameFileName, temp); temp.close(); + m->openOutputFile(scrapNameFileName, temp); temp.close(); + } + + driverCreateTrim(filename, qFileName, (trimFASTAFileName + toString(processors-1) + ".temp"), (scrapFASTAFileName + toString(processors-1) + ".temp"), (trimQualFileName + toString(processors-1) + ".temp"), (scrapQualFileName + toString(processors-1) + ".temp"), (trimNameFileName + toString(processors-1) + ".temp"), (scrapNameFileName + toString(processors-1) + ".temp"), (groupFile + toString(processors-1) + ".temp"), fastaFileNames, qualFileNames, nameFileNames, lines[processors-1], qLines[processors-1]); + processIDS.push_back(processors-1); + + + //Wait until all threads have terminated. + WaitForMultipleObjects(processors-1, hThreadArray, TRUE, INFINITE); + + //Close all thread handles and free memory allocations. + for(int i=0; i < pDataArray.size(); i++){ + for (map::iterator it = pDataArray[i]->groupCounts.begin(); it != pDataArray[i]->groupCounts.end(); it++) { + map::iterator it2 = groupCounts.find(it->first); + if (it2 == groupCounts.end()) { groupCounts[it->first] = it->second; } + else { groupCounts[it->first] += it->second; } + } + CloseHandle(hThreadArray[i]); + delete pDataArray[i]; + } + +#endif + + + //append files for(int i=0;imothurOut("Appending files from process " + toString(processIDS[i])); m->mothurOutEndLine(); m->appendFiles((trimFASTAFileName + toString(processIDS[i]) + ".temp"), trimFASTAFileName); - remove((trimFASTAFileName + toString(processIDS[i]) + ".temp").c_str()); + m->mothurRemove((trimFASTAFileName + toString(processIDS[i]) + ".temp")); m->appendFiles((scrapFASTAFileName + toString(processIDS[i]) + ".temp"), scrapFASTAFileName); - remove((scrapFASTAFileName + toString(processIDS[i]) + ".temp").c_str()); + m->mothurRemove((scrapFASTAFileName + toString(processIDS[i]) + ".temp")); if(qFileName != ""){ m->appendFiles((trimQualFileName + toString(processIDS[i]) + ".temp"), trimQualFileName); - remove((trimQualFileName + toString(processIDS[i]) + ".temp").c_str()); + m->mothurRemove((trimQualFileName + toString(processIDS[i]) + ".temp")); m->appendFiles((scrapQualFileName + toString(processIDS[i]) + ".temp"), scrapQualFileName); - remove((scrapQualFileName + toString(processIDS[i]) + ".temp").c_str()); + m->mothurRemove((scrapQualFileName + toString(processIDS[i]) + ".temp")); } if(nameFile != ""){ m->appendFiles((trimNameFileName + toString(processIDS[i]) + ".temp"), trimNameFileName); - remove((trimNameFileName + toString(processIDS[i]) + ".temp").c_str()); + m->mothurRemove((trimNameFileName + toString(processIDS[i]) + ".temp")); m->appendFiles((scrapNameFileName + toString(processIDS[i]) + ".temp"), scrapNameFileName); - remove((scrapNameFileName + toString(processIDS[i]) + ".temp").c_str()); + m->mothurRemove((scrapNameFileName + toString(processIDS[i]) + ".temp")); } - m->appendFiles((groupFile + toString(processIDS[i]) + ".temp"), groupFile); - remove((groupFile + toString(processIDS[i]) + ".temp").c_str()); + if(createGroup){ + m->appendFiles((groupFile + toString(processIDS[i]) + ".temp"), groupFile); + m->mothurRemove((groupFile + toString(processIDS[i]) + ".temp")); + } if(allFiles){ @@ -852,40 +1056,47 @@ int TrimSeqsCommand::createProcessesCreateTrim(string filename, string qFileName for(int k=0;kappendFiles((fastaFileNames[j][k] + toString(processIDS[i]) + ".temp"), fastaFileNames[j][k]); - remove((fastaFileNames[j][k] + toString(processIDS[i]) + ".temp").c_str()); + m->mothurRemove((fastaFileNames[j][k] + toString(processIDS[i]) + ".temp")); if(qFileName != ""){ m->appendFiles((qualFileNames[j][k] + toString(processIDS[i]) + ".temp"), qualFileNames[j][k]); - remove((qualFileNames[j][k] + toString(processIDS[i]) + ".temp").c_str()); + m->mothurRemove((qualFileNames[j][k] + toString(processIDS[i]) + ".temp")); } if(nameFile != ""){ m->appendFiles((nameFileNames[j][k] + toString(processIDS[i]) + ".temp"), nameFileNames[j][k]); - remove((nameFileNames[j][k] + toString(processIDS[i]) + ".temp").c_str()); + m->mothurRemove((nameFileNames[j][k] + toString(processIDS[i]) + ".temp")); } } } } } - ifstream in; - string tempFile = filename + toString(processIDS[i]) + ".num.temp"; - m->openInputFile(tempFile, in); - int tempNum; - string group; - while (!in.eof()) { - in >> group >> tempNum; m->gobble(in); + #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) || (__linux__) || (__unix__) || (__unix) + if(createGroup){ + ifstream in; + string tempFile = filename + toString(processIDS[i]) + ".num.temp"; + m->openInputFile(tempFile, in); + int tempNum; + string group; + + in >> tempNum; m->gobble(in); - map::iterator it = groupCounts.find(group); - if (it == groupCounts.end()) { groupCounts[group] = tempNum; } - else { groupCounts[it->first] += tempNum; } + if (tempNum != 0) { + while (!in.eof()) { + in >> group >> tempNum; m->gobble(in); + + map::iterator it = groupCounts.find(group); + if (it == groupCounts.end()) { groupCounts[group] = tempNum; } + else { groupCounts[it->first] += tempNum; } + } + } + in.close(); m->mothurRemove(tempFile); } - in.close(); remove(tempFile.c_str()); - + #endif } - - return exitCommand; -#endif + + return exitCommand; } catch(exception& e) { m->errorOut(e, "TrimSeqsCommand", "createProcessesCreateTrim"); @@ -895,14 +1106,16 @@ int TrimSeqsCommand::createProcessesCreateTrim(string filename, string qFileName /**************************************************************************************************/ -int TrimSeqsCommand::setLines(string filename, string qfilename, vector& fastaFilePos, vector& qfileFilePos) { +int TrimSeqsCommand::setLines(string filename, string qfilename) { try { - #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) + + vector fastaFilePos; + vector qfileFilePos; + + #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) || (__linux__) || (__unix__) || (__unix) //set file positions for fasta file fastaFilePos = m->divideFile(filename, processors); - if (qfilename == "") { return processors; } - //get name of first sequence in each chunk map firstSeqNames; for (int i = 0; i < (fastaFilePos.size()-1); i++) { @@ -915,92 +1128,101 @@ int TrimSeqsCommand::setLines(string filename, string qfilename, vectoropenInputFile(qfilename, inQual); - - string input; - while(!inQual.eof()){ - input = m->getline(inQual); - - if (input.length() != 0) { - if(input[0] == '>'){ //this is a sequence name line - istringstream nameStream(input); - - string sname = ""; nameStream >> sname; - sname = sname.substr(1); - - map::iterator it = firstSeqNames.find(sname); - - if(it != firstSeqNames.end()) { //this is the start of a new chunk - unsigned long int pos = inQual.tellg(); - qfileFilePos.push_back(pos - input.length() - 1); - firstSeqNames.erase(it); - } - } - } - - if (firstSeqNames.size() == 0) { break; } - } - inQual.close(); - - - if (firstSeqNames.size() != 0) { - for (map::iterator it = firstSeqNames.begin(); it != firstSeqNames.end(); it++) { - m->mothurOut(it->first + " is in your fasta file and not in your quality file, not using quality file."); m->mothurOutEndLine(); - } - qFileName = ""; - return processors; - } - - //get last file position of qfile - FILE * pFile; - unsigned long int size; - //get num bytes in file - pFile = fopen (qfilename.c_str(),"rb"); - if (pFile==NULL) perror ("Error opening file"); - else{ - fseek (pFile, 0, SEEK_END); - size=ftell (pFile); - fclose (pFile); - } - - qfileFilePos.push_back(size); + if(qfilename != "") { + //seach for filePos of each first name in the qfile and save in qfileFilePos + ifstream inQual; + m->openInputFile(qfilename, inQual); + + string input; + while(!inQual.eof()){ + input = m->getline(inQual); + + if (input.length() != 0) { + if(input[0] == '>'){ //this is a sequence name line + istringstream nameStream(input); + + string sname = ""; nameStream >> sname; + sname = sname.substr(1); + + map::iterator it = firstSeqNames.find(sname); + + if(it != firstSeqNames.end()) { //this is the start of a new chunk + unsigned long long pos = inQual.tellg(); + qfileFilePos.push_back(pos - input.length() - 1); + firstSeqNames.erase(it); + } + } + } + + if (firstSeqNames.size() == 0) { break; } + } + inQual.close(); + + + if (firstSeqNames.size() != 0) { + for (map::iterator it = firstSeqNames.begin(); it != firstSeqNames.end(); it++) { + m->mothurOut(it->first + " is in your fasta file and not in your quality file, not using quality file."); m->mothurOutEndLine(); + } + qFileName = ""; + return processors; + } + + //get last file position of qfile + FILE * pFile; + unsigned long long size; + + //get num bytes in file + pFile = fopen (qfilename.c_str(),"rb"); + if (pFile==NULL) perror ("Error opening file"); + else{ + fseek (pFile, 0, SEEK_END); + size=ftell (pFile); + fclose (pFile); + } + + qfileFilePos.push_back(size); + } + + for (int i = 0; i < (fastaFilePos.size()-1); i++) { + lines.push_back(linePair(fastaFilePos[i], fastaFilePos[(i+1)])); + if (qfilename != "") { qLines.push_back(linePair(qfileFilePos[i], qfileFilePos[(i+1)])); } + } + if(qfilename == "") { qLines = lines; } //files with duds return processors; #else - - fastaFilePos.push_back(0); qfileFilePos.push_back(0); - //get last file position of fastafile - FILE * pFile; - unsigned long int size; - - //get num bytes in file - pFile = fopen (filename.c_str(),"rb"); - if (pFile==NULL) perror ("Error opening file"); - else{ - fseek (pFile, 0, SEEK_END); - size=ftell (pFile); - fclose (pFile); - } - fastaFilePos.push_back(size); - - //get last file position of fastafile - FILE * qFile; - - //get num bytes in file - qFile = fopen (qfilename.c_str(),"rb"); - if (qFile==NULL) perror ("Error opening file"); - else{ - fseek (qFile, 0, SEEK_END); - size=ftell (qFile); - fclose (qFile); - } - qfileFilePos.push_back(size); - + + if (processors == 1) { //save time + //fastaFilePos.push_back(0); qfileFilePos.push_back(0); + //fastaFilePos.push_back(1000); qfileFilePos.push_back(1000); + lines.push_back(linePair(0, 1000)); + if (qfilename != "") { qLines.push_back(linePair(0, 1000)); } + }else{ + int numFastaSeqs = 0; + fastaFilePos = m->setFilePosFasta(filename, numFastaSeqs); + if (fastaFilePos.size() < processors) { processors = fastaFilePos.size(); } + + if (qfilename != "") { + int numQualSeqs = 0; + qfileFilePos = m->setFilePosFasta(qfilename, numQualSeqs); + + if (numFastaSeqs != numQualSeqs) { + m->mothurOut("[ERROR]: You have " + toString(numFastaSeqs) + " sequences in your fasta file, but " + toString(numQualSeqs) + " sequences in your quality file."); m->mothurOutEndLine(); m->control_pressed = true; + } + } + + //figure out how many sequences you have to process + int numSeqsPerProcessor = numFastaSeqs / processors; + for (int i = 0; i < processors; i++) { + int startIndex = i * numSeqsPerProcessor; + if(i == (processors - 1)){ numSeqsPerProcessor = numFastaSeqs - i * numSeqsPerProcessor; } + lines.push_back(linePair(fastaFilePos[startIndex], numSeqsPerProcessor)); + if (qfilename != "") { qLines.push_back(linePair(qfileFilePos[startIndex], numSeqsPerProcessor)); } + } + } + if(qfilename == "") { qLines = lines; } //files with duds return 1; #endif @@ -1013,7 +1235,7 @@ int TrimSeqsCommand::setLines(string filename, string qfilename, vector >& fastaFileNames, vector >& qualFileNames, vector >& nameFileNames){ +bool TrimSeqsCommand::getOligos(vector >& fastaFileNames, vector >& qualFileNames, vector >& nameFileNames){ try { ifstream inOligos; m->openInputFile(oligoFile, inOligos); @@ -1064,19 +1286,46 @@ void TrimSeqsCommand::getOligos(vector >& fastaFileNames, vector< primerNameVector.push_back(group); } else if(type == "REVERSE"){ - Sequence oligoRC("reverse", oligo); - oligoRC.reverseComplement(); - revPrimer.push_back(oligoRC.getUnaligned()); + //Sequence oligoRC("reverse", oligo); + //oligoRC.reverseComplement(); + string oligoRC = reverseOligo(oligo); + revPrimer.push_back(oligoRC); } else if(type == "BARCODE"){ inOligos >> group; + + //barcode lines can look like BARCODE atgcatgc groupName - for 454 seqs + //or BARCODE atgcatgc atgcatgc groupName - for illumina data that has forward and reverse info + string temp = ""; + while (!inOligos.eof()) { + char c = inOligos.get(); + if (c == 10 || c == 13){ break; } + else if (c == 32 || c == 9){;} //space or tab + else { temp += c; } + } + //then this is illumina data with 4 columns + if (temp != "") { + string reverseBarcode = reverseOligo(group); //reverse barcode + group = temp; + + //check for repeat barcodes + map::iterator itBar = rbarcodes.find(reverseBarcode); + if (itBar != rbarcodes.end()) { m->mothurOut("barcode " + reverseBarcode + " is in your oligos file already."); m->mothurOutEndLine(); } + + rbarcodes[reverseBarcode]=indexBarcode; + } + //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]=indexBarcode; indexBarcode++; barcodeNameVector.push_back(group); + }else if(type == "LINKER"){ + linker.push_back(oligo); + }else if(type == "SPACER"){ + spacer.push_back(oligo); } else{ m->mothurOut(type + " is not recognized as a valid type. Choices are forward, reverse, and barcode. Ignoring " + oligo + "."); m->mothurOutEndLine(); } } @@ -1168,286 +1417,37 @@ void TrimSeqsCommand::getOligos(vector >& fastaFileNames, vector< } numFPrimers = primers.size(); numRPrimers = revPrimer.size(); - - } - catch(exception& e) { - m->errorOut(e, "TrimSeqsCommand", "getOligos"); - exit(1); - } -} - -//*************************************************************************************************************** - -int TrimSeqsCommand::stripBarcode(Sequence& seq, QualityScores& qual, int& group){ - try { - - string rawSequence = seq.getUnaligned(); - int success = bdiffs + 1; //guilty until proven innocent + numLinkers = linker.size(); + numSpacers = spacer.size(); - //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 = 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())); - - if(qual.getName() != ""){ - qual.trimQScores(oligo.length(), -1); - } - - success = 0; + bool allBlank = true; + for (int i = 0; i < barcodeNameVector.size(); i++) { + if (barcodeNameVector[i] != "") { + allBlank = false; break; } } - - //if you found the barcode or if you don't want to allow for diffs - if ((bdiffs == 0) || (success == 0)) { return success; } - - else { //try aligning and see if you can find it - - 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 numDiff = countDiffs(oligo, temp); - - 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; } - - } - - return success; - - } - catch(exception& e) { - m->errorOut(e, "TrimSeqsCommand", "stripBarcode"); - exit(1); - } - -} - -//*************************************************************************************************************** - -int TrimSeqsCommand::stripForward(Sequence& seq, QualityScores& qual, int& group){ - try { - string rawSequence = seq.getUnaligned(); - int success = pdiffs + 1; //guilty until proven innocent - - //can you find the primer - for(map::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())); - if(qual.getName() != ""){ - qual.trimQScores(oligo.length(), -1); - } - success = 0; + for (int i = 0; i < primerNameVector.size(); i++) { + if (primerNameVector[i] != "") { + allBlank = false; break; } } - - //if you found the barcode or if you don't want to allow for diffs - if ((pdiffs == 0) || (success == 0)) { return success; } - - else { //try aligning and see if you can find it - - 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 numDiff = countDiffs(oligo, temp); - - 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; } - + if (allBlank) { + m->mothurOut("[WARNING]: your oligos file does not contain any group names. mothur will not create a groupfile."); m->mothurOutEndLine(); + allFiles = false; + return false; } - return success; - - } - catch(exception& e) { - m->errorOut(e, "TrimSeqsCommand", "stripForward"); - exit(1); - } -} - -//*************************************************************************************************************** - -bool TrimSeqsCommand::stripReverse(Sequence& seq, QualityScores& qual){ - try { - string rawSequence = seq.getUnaligned(); - bool success = 0; //guilty until proven innocent - - for(int i=0;ierrorOut(e, "TrimSeqsCommand", "stripReverse"); + m->errorOut(e, "TrimSeqsCommand", "getOligos"); exit(1); } } - //*************************************************************************************************************** bool TrimSeqsCommand::keepFirstTrim(Sequence& sequence, QualityScores& qscores){ @@ -1534,6 +1534,46 @@ bool TrimSeqsCommand::cullHomoP(Sequence& seq){ } } +//********************************************************************/ +string TrimSeqsCommand::reverseOligo(string oligo){ + try { + string reverse = ""; + + for(int i=oligo.length()-1;i>=0;i--){ + + if(oligo[i] == 'A') { reverse += 'T'; } + else if(oligo[i] == 'T'){ reverse += 'A'; } + else if(oligo[i] == 'U'){ reverse += 'A'; } + + else if(oligo[i] == 'G'){ reverse += 'C'; } + else if(oligo[i] == 'C'){ reverse += 'G'; } + + else if(oligo[i] == 'R'){ reverse += 'Y'; } + else if(oligo[i] == 'Y'){ reverse += 'R'; } + + else if(oligo[i] == 'M'){ reverse += 'K'; } + else if(oligo[i] == 'K'){ reverse += 'M'; } + + else if(oligo[i] == 'W'){ reverse += 'W'; } + else if(oligo[i] == 'S'){ reverse += 'S'; } + + else if(oligo[i] == 'B'){ reverse += 'V'; } + else if(oligo[i] == 'V'){ reverse += 'B'; } + + else if(oligo[i] == 'D'){ reverse += 'H'; } + else if(oligo[i] == 'H'){ reverse += 'D'; } + + else { reverse += 'N'; } + } + + + return reverse; + } + catch(exception& e) { + m->errorOut(e, "TrimSeqsCommand", "reverseOligo"); + exit(1); + } +} //*************************************************************************************************************** @@ -1553,80 +1593,4 @@ bool TrimSeqsCommand::cullAmbigs(Sequence& seq){ } } - -//*************************************************************************************************************** - -bool TrimSeqsCommand::compareDNASeq(string oligo, string seq){ - try { - bool success = 1; - int length = oligo.length(); - - for(int i=0;ierrorOut(e, "TrimSeqsCommand", "compareDNASeq"); - exit(1); - } - -} - -//*************************************************************************************************************** - -int TrimSeqsCommand::countDiffs(string oligo, string seq){ - try { - - int length = oligo.length(); - int countDiffs = 0; - - for(int i=0;ierrorOut(e, "TrimSeqsCommand", "countDiffs"); - exit(1); - } - -} - //***************************************************************************************************************