X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=trimseqscommand.cpp;h=cbf5e13603ced805d185196fad9129eacb573d22;hb=bc4c4d63e983a3e91e74b5038bfec6705e1a3a2e;hp=71e5c9277af61fef9884f42b2c8e9d54a2796e12;hpb=a0f87c2ae6414af28d4e70b1e6830401eac21bef;p=mothur.git diff --git a/trimseqscommand.cpp b/trimseqscommand.cpp index 71e5c92..cbf5e13 100644 --- a/trimseqscommand.cpp +++ b/trimseqscommand.cpp @@ -16,6 +16,7 @@ TrimSeqsCommand::TrimSeqsCommand(string option) { try { abort = false; + comboStarts = 0; //allow user to run help if(option == "help") { help(); abort = true; } @@ -224,8 +225,9 @@ 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)); @@ -266,8 +268,9 @@ int TrimSeqsCommand::execute(){ if (m->control_pressed) { return 0; } #else 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)); @@ -279,22 +282,37 @@ int TrimSeqsCommand::execute(){ for(int i=0;i'){ - inFASTA >> seqName; - outGroups << seqName << '\t' << groupVector[i] << endl; + 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(getRootName(fastaFile) + groupVector[i] + ".fasta", inFASTA); + openInputFile(fastaFileNames[i], inFASTA); + ofstream outGroups; + string outGroupFilename = outputDir + getRootName(getSimpleName(fastaFileNames[i])) + "groups"; + openOutputFile(outGroupFilename, outGroups); + //openOutputFile(outputDir + getRootName(getSimpleName(fastaFile)) + groupVector[i] + ".groups", 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; } } } - while (!inFASTA.eof()) { char c = inFASTA.get(); if (c == 10 || c == 13){ break; } } + outGroups.close(); + inFASTA.close(); } - outGroups.close(); - inFASTA.close(); } if (m->control_pressed) { @@ -351,15 +369,8 @@ int TrimSeqsCommand::driverCreateTrim(string filename, string qFileName, string for(int i=0;inum;i++){ if (m->control_pressed) { - inFASTA.close(); - outFASTA.close(); - scrapFASTA.close(); - if (oligoFile != "") { outGroups.close(); } - if(qFileName != "") { qFile.close(); } - for(int i=0;iclose(); - delete fastaFileNames[i]; - } + 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; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } return 0; } @@ -370,7 +381,7 @@ int TrimSeqsCommand::driverCreateTrim(string filename, string qFileName, string string origSeq = currSeq.getUnaligned(); if (origSeq != "") { - int group; + int groupBar, groupPrime; string trashCode = ""; int currentSeqsDiffs = 0; @@ -386,13 +397,13 @@ int TrimSeqsCommand::driverCreateTrim(string filename, string qFileName, string } if(barcodes.size() != 0){ - success = stripBarcode(currSeq, group); + success = stripBarcode(currSeq, groupBar); if(success > bdiffs){ trashCode += 'b'; } else{ currentSeqsDiffs += success; } } if(numFPrimers != 0){ - success = stripForward(currSeq); + success = stripForward(currSeq, groupPrime); if(success > pdiffs){ trashCode += 'f'; } else{ currentSeqsDiffs += success; } } @@ -423,10 +434,18 @@ int TrimSeqsCommand::driverCreateTrim(string filename, string qFileName, string currSeq.setAligned(currSeq.getUnaligned()); currSeq.printSequence(outFASTA); 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]); } } } @@ -560,14 +579,18 @@ 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)) + toString(index) + "." + group + ".fasta")); + 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)) + toString(index) + "." + group + ".fasta")); + }else { + outputNames.push_back((outputDir + getRootName(getSimpleName(fastaFile)) + toString(index) + "." + group + ".fasta")); + } + } + } - 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")); + outputNames.push_back((outputDir + getRootName(getSimpleName(fastaFile)) + toString(index) + "." + group + ".fasta")); + outFASTAVec.push_back((outputDir + getRootName(getSimpleName(fastaFile)) + toString(index) + "." + group + ".fasta")); } - } + }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(fastaFile)) + toString(itBar->second) + "." + groupVector[itBar->second] + "." + toString(itPrime->second) + "." + groupVector[itPrime->second] + ".fasta")); + outFASTAVec.push_back((outputDir + getRootName(getSimpleName(fastaFile)) + toString(itBar->second) + "." + groupVector[itBar->second] + "." + toString(itPrime->second) + "." + groupVector[itPrime->second] + ".fasta")); + combos[(groupVector[itBar->second] + "." + groupVector[itPrime->second])] = outFASTAVec.size()-1; + } + } + } + } + + numFPrimers = primers.size(); numRPrimers = revPrimer.size(); } @@ -650,7 +717,7 @@ int TrimSeqsCommand::stripBarcode(Sequence& seq, int& group){ maxLength = it->first.length(); } } - alignment = new NeedlemanOverlap(-2.0, 1.0, -1.0, (maxLength+bdiffs+1)); + alignment = new NeedlemanOverlap(-1.0, 1.0, -1.0, (maxLength+bdiffs+1)); }else{ alignment = NULL; } @@ -684,6 +751,9 @@ int TrimSeqsCommand::stripBarcode(Sequence& seq, int& group){ int newStart=0; int numDiff = countDiffs(oligo, temp); + +// cout << oligo << '\t' << temp << '\t' << numDiff << endl; + if(numDiff < minDiff){ minDiff = numDiff; minCount = 1; @@ -726,27 +796,27 @@ int TrimSeqsCommand::stripBarcode(Sequence& seq, int& group){ //*************************************************************************************************************** -int TrimSeqsCommand::stripForward(Sequence& seq){ +int TrimSeqsCommand::stripForward(Sequence& seq, int& group){ try { string rawSequence = seq.getUnaligned(); int success = pdiffs + 1; //guilty until proven innocent //can you find the primer - 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 = 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; } @@ -757,24 +827,27 @@ int TrimSeqsCommand::stripForward(Sequence& seq){ int maxLength = 0; Alignment* alignment; - if (numFPrimers > 0) { + if (primers.size() > 0) { + map::iterator it=primers.begin(); - for(int i=0;i maxLength){ - maxLength = forPrimer[i].length(); + for(it;it!=primers.end();it++){ + if(it->first.length() > maxLength){ + maxLength = it->first.length(); } } - alignment = new NeedlemanOverlap(-2.0, 1.0, -1.0, (maxLength+pdiffs+1)); + 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(int i=0;i::iterator it=primers.begin();it!=primers.end();it++){ + string oligo = it->first; +// int length = oligo.length(); if(rawSequence.length() < maxLength){ success = pdiffs + 100; @@ -793,12 +866,16 @@ int TrimSeqsCommand::stripForward(Sequence& seq){ } 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; } - else if(minCount > 1) { success = pdiffs + 10; } - else{ + + if(minDiff > 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)); success = minDiff; } @@ -821,6 +900,7 @@ int TrimSeqsCommand::stripForward(Sequence& seq){ if (alignment != NULL) { delete alignment; } } + return success; }