X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=trimoligos.cpp;h=a53f4e85388f0f7bbb6fda40c2074b4f555e29d3;hb=01d5a60fc5f396339acf529151fa8186ee7c1a41;hp=2f92cc847ca75e1a0983aeef2d09e9eaab68e57f;hpb=e0ce7cbc93d7d2fbb753ca694182db092a0ea0e7;p=mothur.git diff --git a/trimoligos.cpp b/trimoligos.cpp index 2f92cc8..a53f4e8 100644 --- a/trimoligos.cpp +++ b/trimoligos.cpp @@ -14,7 +14,7 @@ /********************************************************************/ //strip, pdiffs, bdiffs, primers, barcodes, revPrimers -TrimOligos::TrimOligos(int p, int b, int l, int s, map pr, map br, map rbr, vector r, vector lk, vector sp){ +TrimOligos::TrimOligos(int p, int b, int l, int s, map pr, map br, vector r, vector lk, vector sp){ try { m = MothurOut::getInstance(); @@ -24,11 +24,37 @@ TrimOligos::TrimOligos(int p, int b, int l, int s, map pr, map::iterator it=barcodes.begin();it!=barcodes.end();it++){ + if(it->first.length() > maxFBarcodeLength){ + maxFBarcodeLength = it->first.length(); + } + } + maxFPrimerLength = 0; + map::iterator it; + for(it=primers.begin();it!=primers.end();it++){ + if(it->first.length() > maxFPrimerLength){ + maxFPrimerLength = it->first.length(); + } + } + + maxLinkerLength = 0; + for(int i = 0; i < linker.size(); i++){ + if(linker[i].length() > maxLinkerLength){ + maxLinkerLength = linker[i].length(); + } + } + + maxSpacerLength = 0; + for(int i = 0; i < spacer.size(); i++){ + if(spacer[i].length() > maxSpacerLength){ + maxSpacerLength = spacer[i].length(); + } + } } catch(exception& e) { m->errorOut(e, "TrimOligos", "TrimOligos"); @@ -37,7 +63,7 @@ TrimOligos::TrimOligos(int p, int b, int l, int s, map pr, map pr, map br, vector r, vector lk, vector sp){ +TrimOligos::TrimOligos(int p, int b, int l, int s, map pr, map br){ try { m = MothurOut::getInstance(); @@ -46,11 +72,60 @@ TrimOligos::TrimOligos(int p, int b, int l, int s, map pr, map::iterator it=br.begin();it!=br.end();it++){ + string forward = it->second.forward; + map >::iterator itForward = ifbarcodes.find(forward); + + if(forward.length() > maxFBarcodeLength){ maxFBarcodeLength = forward.length(); } + + if (itForward == ifbarcodes.end()) { + vector temp; temp.push_back(it->first); + ifbarcodes[forward] = temp; + }else { itForward->second.push_back(it->first); } + } + + maxFPrimerLength = 0; + for(map::iterator it=pr.begin();it!=pr.end();it++){ + string forward = it->second.forward; + map >::iterator itForward = ifprimers.find(forward); + + if(forward.length() > maxFPrimerLength){ maxFPrimerLength = forward.length(); } + + if (itForward == ifprimers.end()) { + vector temp; temp.push_back(it->first); + ifprimers[forward] = temp; + }else { itForward->second.push_back(it->first); } + } + + maxRBarcodeLength = 0; + for(map::iterator it=br.begin();it!=br.end();it++){ + string forward = it->second.reverse; + map >::iterator itForward = irbarcodes.find(forward); + + if(forward.length() > maxRBarcodeLength){ maxRBarcodeLength = forward.length(); } + + if (itForward == irbarcodes.end()) { + vector temp; temp.push_back(it->first); + irbarcodes[forward] = temp; + }else { itForward->second.push_back(it->first); } + } + + maxRPrimerLength = 0; + for(map::iterator it=pr.begin();it!=pr.end();it++){ + string forward = it->second.reverse; + map >::iterator itForward = irprimers.find(forward); + + if(forward.length() > maxRPrimerLength){ maxRPrimerLength = forward.length(); } + + if (itForward == irprimers.end()) { + vector temp; temp.push_back(it->first); + irprimers[forward] = temp; + }else { itForward->second.push_back(it->first); } + } + + ipbarcodes = br; + ipprimers = pr; } catch(exception& e) { m->errorOut(e, "TrimOligos", "TrimOligos"); @@ -69,6 +144,24 @@ TrimOligos::TrimOligos(int p, int b, map pr, map br, v barcodes = br; primers = pr; revPrimer = r; + + maxFBarcodeLength = 0; + for(map::iterator it=barcodes.begin();it!=barcodes.end();it++){ + string oligo = it->first; + if(oligo.length() > maxFBarcodeLength){ + maxFBarcodeLength = oligo.length(); + } + } + maxRBarcodeLength = maxFBarcodeLength; + + maxFPrimerLength = 0; + for(map::iterator it=primers.begin();it!=primers.end();it++){ + string oligo = it->first; + if(oligo.length() > maxFPrimerLength){ + maxFPrimerLength = oligo.length(); + } + } + maxRPrimerLength = maxFPrimerLength; } catch(exception& e) { m->errorOut(e, "TrimOligos", "TrimOligos"); @@ -77,7 +170,135 @@ TrimOligos::TrimOligos(int p, int b, map pr, map br, v } /********************************************************************/ TrimOligos::~TrimOligos() {} +//********************************************************************/ +bool TrimOligos::findForward(Sequence& seq, int& primerStart, int& primerEnd){ + try { + + string rawSequence = seq.getUnaligned(); + + for(map::iterator it=primers.begin();it!=primers.end();it++){ + string oligo = it->first; + + if(rawSequence.length() < oligo.length()) { break; } + + //search for primer + int olength = oligo.length(); + for (int j = 0; j < rawSequence.length()-olength; j++){ + if (m->control_pressed) { primerStart = 0; primerEnd = 0; return false; } + string rawChunk = rawSequence.substr(j, olength); + if(compareDNASeq(oligo, rawChunk)) { + primerStart = j; + primerEnd = primerStart + olength; + return true; + } + + } + } + + primerStart = 0; primerEnd = 0; + //if you don't want to allow for diffs + if (pdiffs == 0) { return false; } + else { //try aligning and see if you can find it + + //can you find the barcode + int minDiff = 1e6; + int minCount = 1; + + Alignment* alignment; + if (primers.size() > 0) { alignment = new NeedlemanOverlap(-1.0, 1.0, -1.0, (maxFPrimerLength+pdiffs+1)); } + else{ alignment = NULL; } + + for(map::iterator it=primers.begin();it!=primers.end();it++){ + + string prim = it->first; + //search for primer + int olength = prim.length(); + if (rawSequence.length() < olength) {} //ignore primers too long for this seq + else{ + for (int j = 0; j < rawSequence.length()-olength; j++){ + + string oligo = it->first; + + if (m->control_pressed) { primerStart = 0; primerEnd = 0; return false; } + + string rawChunk = rawSequence.substr(j, olength+pdiffs); + + //use needleman to align first primer.length()+numdiffs of sequence to each barcode + alignment->align(oligo, rawChunk); + 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; + primerStart = j; + primerEnd = primerStart + alnLength; + }else if(numDiff == minDiff){ minCount++; } + } + } + } + + if (alignment != NULL) { delete alignment; } + + if(minDiff > pdiffs) { primerStart = 0; primerEnd = 0; return false; } //no good matches + else if(minCount > 1) { primerStart = 0; primerEnd = 0; return false; } //can't tell the difference between multiple primers + else{ return true; } + } + + primerStart = 0; primerEnd = 0; + return false; + + } + catch(exception& e) { + m->errorOut(e, "TrimOligos", "stripForward"); + exit(1); + } +} +//******************************************************************/ +bool TrimOligos::findReverse(Sequence& seq, int& primerStart, int& primerEnd){ + try { + + string rawSequence = seq.getUnaligned(); + + for(int i=0;i= 0; j--){ + if (m->control_pressed) { primerStart = 0; primerEnd = 0; return false; } + string rawChunk = rawSequence.substr(j, olength); + + if(compareDNASeq(oligo, rawChunk)) { + primerStart = j; + primerEnd = primerStart + olength; + return true; + } + + } + } + + primerStart = 0; primerEnd = 0; + return false; + } + catch(exception& e) { + m->errorOut(e, "PcrSeqsCommand", "findReverse"); + exit(1); + } +} //*******************************************************************/ + int TrimOligos::stripBarcode(Sequence& seq, QualityScores& qual, int& group){ try { @@ -109,21 +330,9 @@ int TrimOligos::stripBarcode(Sequence& seq, QualityScores& qual, int& group){ 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; - - for(it=barcodes.begin();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; } + if (barcodes.size() > 0) {alignment = new NeedlemanOverlap(-1.0, 1.0, -1.0, (maxFBarcodeLength+bdiffs+1)); } + else{ alignment = NULL; } //can you find the barcode int minDiff = 1e6; @@ -135,7 +344,7 @@ int TrimOligos::stripBarcode(Sequence& seq, QualityScores& qual, int& group){ string oligo = it->first; // int length = oligo.length(); - if(rawSequence.length() < maxLength){ //let's just assume that the barcodes are the same length + if(rawSequence.length() < maxFBarcodeLength){ //let's just assume that the barcodes are the same length success = bdiffs + 10; break; } @@ -194,27 +403,33 @@ int TrimOligos::stripBarcode(Sequence& seq, QualityScores& qual, int& group){ m->errorOut(e, "TrimOligos", "stripBarcode"); exit(1); } - } //*******************************************************************/ -int TrimOligos::stripBarcode(Sequence& seq, int& group){ +int TrimOligos::stripBarcode(Sequence& forwardSeq, Sequence& reverseSeq, int& group){ try { - - string rawSequence = seq.getUnaligned(); + //look for forward barcode + string rawFSequence = forwardSeq.getUnaligned(); + string rawRSequence = reverseSeq.getUnaligned(); 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 + //can you find the forward barcode + for(map::iterator it=ipbarcodes.begin();it!=ipbarcodes.end();it++){ + string foligo = it->second.forward; + string roligo = it->second.reverse; + + if(rawFSequence.length() < foligo.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(rawRSequence.length() < roligo.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((compareDNASeq(foligo, rawFSequence.substr(0,foligo.length()))) && (compareDNASeq(roligo, rawRSequence.substr((rawRSequence.length()-roligo.length()),roligo.length())))) { + group = it->first; + forwardSeq.setUnaligned(rawFSequence.substr(foligo.length())); + reverseSeq.setUnaligned(rawRSequence.substr(0,(rawRSequence.length()-roligo.length()))); success = 0; break; } @@ -222,230 +437,899 @@ int TrimOligos::stripBarcode(Sequence& seq, int& group){ //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(map::iterator it=barcodes.begin();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; } + if (ifbarcodes.size() > 0) { alignment = new NeedlemanOverlap(-1.0, 1.0, -1.0, (maxFBarcodeLength+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++){ + vector< vector > minFGroup; + vector minFPos; + + //the pair can have duplicates, but we only want to search for the unique forward and reverses, example + /* + 1 Sarah Westcott + 2 John Westcott + 3 Anna Westcott + 4 Sarah Schloss + 5 Pat Schloss + 6 Gail Brown + 7 Pat Moore + + only want to look for forward = Sarah, John, Anna, Pat, Gail + reverse = Westcott, Schloss, Brown, Moore + + but if best match forward = 4, and reverse = 1, we want to count as a valid match because forward 1 and forward 4 are the same. so both barcodes map to same group. + */ + //cout << endl << forwardSeq.getName() << endl; + for(map >::iterator it=ifbarcodes.begin();it!=ifbarcodes.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 + if(rawFSequence.length() < maxFBarcodeLength){ //let's just assume that the barcodes are the same length success = bdiffs + 10; break; } - + //cout << "before = " << oligo << '\t' << rawFSequence.substr(0,oligo.length()+bdiffs) << endl; //use needleman to align first barcode.length()+numdiffs of sequence to each barcode - alignment->align(oligo, rawSequence.substr(0,oligo.length()+bdiffs)); + alignment->align(oligo, rawFSequence.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; } - } + 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); + int numDiff = countDiffs(oligo, temp); + if (alnLength == 0) { numDiff = bdiffs + 100; } + //cout << "after = " << oligo << '\t' << temp << '\t' << numDiff << endl; + if(numDiff < minDiff){ minDiff = numDiff; minCount = 1; - minGroup = it->second; - minPos = 0; + minFGroup.clear(); + minFGroup.push_back(it->second); + int tempminFPos = 0; + minFPos.clear(); for(int i=0;isecond); + int tempminFPos = 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)); - success = minDiff; + else{ + //check for reverse match + if (alignment != NULL) { delete alignment; } + + if (irbarcodes.size() > 0) { alignment = new NeedlemanOverlap(-1.0, 1.0, -1.0, (maxRBarcodeLength+bdiffs+1)); } + else{ alignment = NULL; } + + //can you find the barcode + minDiff = 1e6; + minCount = 1; + vector< vector > minRGroup; + vector minRPos; + + for(map >::iterator it=irbarcodes.begin();it!=irbarcodes.end();it++){ + string oligo = it->first; + //cout << "before = " << oligo << '\t' << rawRSequence.substr(0,oligo.length()+bdiffs) << endl; + if(rawRSequence.length() < maxRBarcodeLength){ //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, rawRSequence.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 (alnLength == 0) { numDiff = bdiffs + 100; } + + //cout << "after = " << oligo << '\t' << temp << '\t' << numDiff << endl; + if(numDiff < minDiff){ + minDiff = numDiff; + minCount = 1; + minRGroup.clear(); + minRGroup.push_back(it->second); + int tempminRPos = 0; + minRPos.clear(); + for(int i=0;isecond); + } + + } + + /*cout << minDiff << '\t' << minCount << '\t' << endl; + for (int i = 0; i < minFGroup.size(); i++) { + cout << i << '\t'; + for (int j = 0; j < minFGroup[i].size(); j++) { cout << minFGroup[i][j] << " "; } + cout << endl; + } + cout << endl; + for (int i = 0; i < minRGroup.size(); i++) { + cout << i << '\t'; + for (int j = 0; j < minRGroup[i].size(); j++) { cout << minRGroup[i][j] << " "; } + cout << endl; + } + cout << endl;*/ + if(minDiff > bdiffs) { success = minDiff; } //no good matches + else { + bool foundMatch = false; + vector matches; + for (int i = 0; i < minFGroup.size(); i++) { + for (int j = 0; j < minFGroup[i].size(); j++) { + for (int k = 0; k < minRGroup.size(); k++) { + if (m->inUsersGroups(minFGroup[i][j], minRGroup[k])) { matches.push_back(minFGroup[i][j]); k+= minRGroup.size(); } + } + } + } + + int fStart = 0; + int rStart = 0; + if (matches.size() == 1) { + foundMatch = true; + group = matches[0]; + fStart = minFPos[0]; + rStart = minRPos[0]; + } + + //we have an acceptable match for the forward and reverse, but do they match? + if (foundMatch) { + forwardSeq.setUnaligned(rawFSequence.substr(fStart)); + reverseSeq.setUnaligned(rawRSequence.substr(rStart)); + success = minDiff; + }else { success = bdiffs + 100; } + } } if (alignment != NULL) { delete alignment; } - } return success; } catch(exception& e) { - m->errorOut(e, "TrimOligos", "stripBarcode"); + m->errorOut(e, "TrimOligos", "stripIBarcode"); exit(1); } } //*******************************************************************/ -int TrimOligos::stripRBarcode(Sequence& seq, QualityScores& qual, int& group){ +int TrimOligos::stripBarcode(Sequence& forwardSeq, Sequence& reverseSeq, QualityScores& forwardQual, QualityScores& reverseQual, int& group){ try { - - string rawSequence = seq.getUnaligned(); + //look for forward barcode + string rawFSequence = forwardSeq.getUnaligned(); + string rawRSequence = reverseSeq.getUnaligned(); int success = bdiffs + 1; //guilty until proven innocent - //can you find the barcode - for(map::iterator it=rbarcodes.begin();it!=rbarcodes.end();it++){ - string oligo = it->first; - if(rawSequence.length() < oligo.length()){ //let's just assume that the barcodes are the same length + //can you find the forward barcode + for(map::iterator it=ipbarcodes.begin();it!=ipbarcodes.end();it++){ + string foligo = it->second.forward; + string roligo = it->second.reverse; + + if(rawFSequence.length() < foligo.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(rawRSequence.length() < roligo.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(foligo, rawFSequence.substr(0,foligo.length()))) && (compareDNASeq(roligo, rawRSequence.substr((rawRSequence.length()-roligo.length()),roligo.length())))) { + group = it->first; + forwardSeq.setUnaligned(rawFSequence.substr(foligo.length())); + reverseSeq.setUnaligned(rawRSequence.substr(0,(rawRSequence.length()-roligo.length()))); + forwardQual.trimQScores(foligo.length(), -1); + reverseQual.trimQScores(-1, rawRSequence.length()-roligo.length()); + success = 0; + 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 + Alignment* alignment; + if (ifbarcodes.size() > 0) { alignment = new NeedlemanOverlap(-1.0, 1.0, -1.0, (maxFBarcodeLength+bdiffs+1)); } + else{ alignment = NULL; } + + //can you find the barcode + int minDiff = 1e6; + int minCount = 1; + vector< vector > minFGroup; + vector minFPos; - if(compareDNASeq(oligo, rawSequence.substr((rawSequence.length()-oligo.length()),oligo.length()))){ - group = it->second; - seq.setUnaligned(rawSequence.substr(0,(rawSequence.length()-oligo.length()))); + //the pair can have duplicates, but we only want to search for the unique forward and reverses, example + /* + 1 Sarah Westcott + 2 John Westcott + 3 Anna Westcott + 4 Sarah Schloss + 5 Pat Schloss + 6 Gail Brown + 7 Pat Moore + + only want to look for forward = Sarah, John, Anna, Pat, Gail + reverse = Westcott, Schloss, Brown, Moore + + but if best match forward = 4, and reverse = 1, we want to count as a valid match because forward 1 and forward 4 are the same. so both barcodes map to same group. + */ + //cout << endl << forwardSeq.getName() << endl; + for(map >::iterator it=ifbarcodes.begin();it!=ifbarcodes.end();it++){ + string oligo = it->first; - if(qual.getName() != ""){ - qual.trimQScores(-1, rawSequence.length()-oligo.length()); + if(rawFSequence.length() < maxFBarcodeLength){ //let's just assume that the barcodes are the same length + success = bdiffs + 10; + break; } + //cout << "before = " << oligo << '\t' << rawFSequence.substr(0,oligo.length()+bdiffs) << endl; + //use needleman to align first barcode.length()+numdiffs of sequence to each barcode + alignment->align(oligo, rawFSequence.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 (alnLength == 0) { numDiff = bdiffs + 100; } + //cout << "after = " << oligo << '\t' << temp << '\t' << numDiff << endl; + + if(numDiff < minDiff){ + minDiff = numDiff; + minCount = 1; + minFGroup.clear(); + minFGroup.push_back(it->second); + int tempminFPos = 0; + minFPos.clear(); + for(int i=0;isecond); + int tempminFPos = 0; + for(int i=0;i bdiffs) { success = minDiff; } //no good matches + else{ + //check for reverse match + if (alignment != NULL) { delete alignment; } + + if (irbarcodes.size() > 0) { alignment = new NeedlemanOverlap(-1.0, 1.0, -1.0, (maxRBarcodeLength+bdiffs+1)); } + else{ alignment = NULL; } + + //can you find the barcode + minDiff = 1e6; + minCount = 1; + vector< vector > minRGroup; + vector minRPos; + + for(map >::iterator it=irbarcodes.begin();it!=irbarcodes.end();it++){ + string oligo = it->first; + //cout << "before = " << oligo << '\t' << rawRSequence.substr(0,oligo.length()+bdiffs) << endl; + if(rawRSequence.length() < maxRBarcodeLength){ //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, rawRSequence.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 (alnLength == 0) { numDiff = bdiffs + 100; } + + //cout << "after = " << oligo << '\t' << temp << '\t' << numDiff << endl; + if(numDiff < minDiff){ + minDiff = numDiff; + minCount = 1; + minRGroup.clear(); + minRGroup.push_back(it->second); + int tempminRPos = 0; + minRPos.clear(); + for(int i=0;isecond); + } + + } + + /*cout << minDiff << '\t' << minCount << '\t' << endl; + for (int i = 0; i < minFGroup.size(); i++) { + cout << i << '\t'; + for (int j = 0; j < minFGroup[i].size(); j++) { cout << minFGroup[i][j] << " "; } + cout << endl; + } + cout << endl; + for (int i = 0; i < minRGroup.size(); i++) { + cout << i << '\t'; + for (int j = 0; j < minRGroup[i].size(); j++) { cout << minRGroup[i][j] << " "; } + cout << endl; + } + cout << endl;*/ + if(minDiff > bdiffs) { success = minDiff; } //no good matches + else { + bool foundMatch = false; + vector matches; + for (int i = 0; i < minFGroup.size(); i++) { + for (int j = 0; j < minFGroup[i].size(); j++) { + for (int k = 0; k < minRGroup.size(); k++) { + if (m->inUsersGroups(minFGroup[i][j], minRGroup[k])) { matches.push_back(minFGroup[i][j]); k+= minRGroup.size(); } + } + } + } + + int fStart = 0; + int rStart = 0; + if (matches.size() == 1) { + foundMatch = true; + group = matches[0]; + fStart = minFPos[0]; + rStart = minRPos[0]; + } + + //we have an acceptable match for the forward and reverse, but do they match? + if (foundMatch) { + forwardSeq.setUnaligned(rawFSequence.substr(fStart)); + reverseSeq.setUnaligned(rawRSequence.substr(rStart)); + forwardQual.trimQScores(fStart, -1); + reverseQual.trimQScores(rStart, -1); + success = minDiff; + }else { success = bdiffs + 100; } + } + } + + if (alignment != NULL) { delete alignment; } + } + + return success; + + } + catch(exception& e) { + m->errorOut(e, "TrimOligos", "stripIBarcode"); + exit(1); + } + +} +//*******************************************************************/ +int TrimOligos::stripForward(Sequence& forwardSeq, Sequence& reverseSeq, QualityScores& forwardQual, QualityScores& reverseQual, int& group){ + try { + //look for forward barcode + string rawFSequence = forwardSeq.getUnaligned(); + string rawRSequence = reverseSeq.getUnaligned(); + int success = pdiffs + 1; //guilty until proven innocent + + //can you find the forward barcode + for(map::iterator it=ipprimers.begin();it!=ipprimers.end();it++){ + string foligo = it->second.forward; + string roligo = it->second.reverse; + + if(rawFSequence.length() < foligo.length()){ //let's just assume that the barcodes are the same length + success = pdiffs + 10; //if the sequence is shorter than the barcode then bail out + break; + } + if(rawRSequence.length() < roligo.length()){ //let's just assume that the barcodes are the same length + success = pdiffs + 10; //if the sequence is shorter than the barcode then bail out + break; + } + + if((compareDNASeq(foligo, rawFSequence.substr(0,foligo.length()))) && (compareDNASeq(roligo, rawRSequence.substr((rawRSequence.length()-roligo.length()),roligo.length())))) { + group = it->first; + forwardSeq.setUnaligned(rawFSequence.substr(foligo.length())); + reverseSeq.setUnaligned(rawRSequence.substr(0,(rawRSequence.length()-roligo.length()))); + forwardQual.trimQScores(foligo.length(), -1); + reverseQual.trimQScores(-1, rawRSequence.length()-roligo.length()); success = 0; break; } } //if you found the barcode or if you don't want to allow for diffs - if ((bdiffs == 0) || (success == 0)) { return success; } - + if ((pdiffs == 0) || (success == 0)) { return success; } else { //try aligning and see if you can find it - - int maxLength = 0; - Alignment* alignment; - if (rbarcodes.size() > 0) { - map::iterator it; + if (ifprimers.size() > 0) { alignment = new NeedlemanOverlap(-1.0, 1.0, -1.0, (maxFPrimerLength+pdiffs+1)); } + else{ alignment = NULL; } + + //can you find the barcode + int minDiff = 1e6; + int minCount = 1; + vector< vector > minFGroup; + vector minFPos; + + //the pair can have duplicates, but we only want to search for the unique forward and reverses, example + /* + 1 Sarah Westcott + 2 John Westcott + 3 Anna Westcott + 4 Sarah Schloss + 5 Pat Schloss + 6 Gail Brown + 7 Pat Moore + + only want to look for forward = Sarah, John, Anna, Pat, Gail + reverse = Westcott, Schloss, Brown, Moore + + but if best match forward = 4, and reverse = 1, we want to count as a valid match because forward 1 and forward 4 are the same. so both barcodes map to same group. + */ + //cout << endl << forwardSeq.getName() << endl; + for(map >::iterator it=ifprimers.begin();it!=ifprimers.end();it++){ + string oligo = it->first; - for(it=rbarcodes.begin();it!=rbarcodes.end();it++){ - if(it->first.length() > maxLength){ - maxLength = it->first.length(); - } + if(rawFSequence.length() < maxFPrimerLength){ //let's just assume that the barcodes are the same length + success = pdiffs + 10; + break; } - alignment = new NeedlemanOverlap(-1.0, 1.0, -1.0, (maxLength+bdiffs+1)); + //cout << "before = " << oligo << '\t' << rawFSequence.substr(0,oligo.length()+pdiffs) << endl; + //use needleman to align first barcode.length()+numdiffs of sequence to each barcode + alignment->align(oligo, rawFSequence.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); - }else{ alignment = NULL; } + if (alnLength == 0) { numDiff = pdiffs + 100; } + //cout << "after = " << oligo << '\t' << temp << '\t' << numDiff << endl; + + if(numDiff < minDiff){ + minDiff = numDiff; + minCount = 1; + minFGroup.clear(); + minFGroup.push_back(it->second); + int tempminFPos = 0; + minFPos.clear(); + for(int i=0;isecond); + int tempminFPos = 0; + for(int i=0;i pdiffs) { success = minDiff; } //no good matches + else{ + //check for reverse match + if (alignment != NULL) { delete alignment; } + + if (irbarcodes.size() > 0) { alignment = new NeedlemanOverlap(-1.0, 1.0, -1.0, (maxRPrimerLength+pdiffs+1)); } + else{ alignment = NULL; } + + //can you find the barcode + minDiff = 1e6; + minCount = 1; + vector< vector > minRGroup; + vector minRPos; + + for(map >::iterator it=irprimers.begin();it!=irprimers.end();it++){ + string oligo = it->first; + //cout << "before = " << oligo << '\t' << rawRSequence.substr(0,oligo.length()+pdiffs) << endl; + if(rawRSequence.length() < maxRPrimerLength){ //let's just assume that the barcodes are the same length + success = pdiffs + 10; + break; + } + + //use needleman to align first barcode.length()+numdiffs of sequence to each barcode + alignment->align(oligo, rawRSequence.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 (alnLength == 0) { numDiff = pdiffs + 100; } + + //cout << "after = " << oligo << '\t' << temp << '\t' << numDiff << endl; + if(numDiff < minDiff){ + minDiff = numDiff; + minCount = 1; + minRGroup.clear(); + minRGroup.push_back(it->second); + int tempminRPos = 0; + minRPos.clear(); + for(int i=0;isecond); + } + + } + + /*cout << minDiff << '\t' << minCount << '\t' << endl; + for (int i = 0; i < minFGroup.size(); i++) { + cout << i << '\t'; + for (int j = 0; j < minFGroup[i].size(); j++) { cout << minFGroup[i][j] << " "; } + cout << endl; + } + cout << endl; + for (int i = 0; i < minRGroup.size(); i++) { + cout << i << '\t'; + for (int j = 0; j < minRGroup[i].size(); j++) { cout << minRGroup[i][j] << " "; } + cout << endl; + } + cout << endl;*/ + if(minDiff > pdiffs) { success = minDiff; } //no good matches + else { + bool foundMatch = false; + vector matches; + for (int i = 0; i < minFGroup.size(); i++) { + for (int j = 0; j < minFGroup[i].size(); j++) { + for (int k = 0; k < minRGroup.size(); k++) { + if (m->inUsersGroups(minFGroup[i][j], minRGroup[k])) { matches.push_back(minFGroup[i][j]); k+= minRGroup.size(); } + } + } + } + + int fStart = 0; + int rStart = 0; + if (matches.size() == 1) { + foundMatch = true; + group = matches[0]; + fStart = minFPos[0]; + rStart = minRPos[0]; + } + + //we have an acceptable match for the forward and reverse, but do they match? + if (foundMatch) { + forwardSeq.setUnaligned(rawFSequence.substr(fStart)); + reverseSeq.setUnaligned(rawRSequence.substr(rStart)); + forwardQual.trimQScores(fStart, -1); + reverseQual.trimQScores(rStart, -1); + success = minDiff; + }else { success = pdiffs + 100; } + } + } + + if (alignment != NULL) { delete alignment; } + } + + return success; + + } + catch(exception& e) { + m->errorOut(e, "TrimOligos", "stripIForward"); + exit(1); + } + +} +//*******************************************************************/ +int TrimOligos::stripForward(Sequence& forwardSeq, Sequence& reverseSeq, int& group){ + try { + //look for forward barcode + string rawFSequence = forwardSeq.getUnaligned(); + string rawRSequence = reverseSeq.getUnaligned(); + int success = pdiffs + 1; //guilty until proven innocent + + //can you find the forward barcode + for(map::iterator it=ipprimers.begin();it!=ipprimers.end();it++){ + string foligo = it->second.forward; + string roligo = it->second.reverse; + + if(rawFSequence.length() < foligo.length()){ //let's just assume that the barcodes are the same length + success = pdiffs + 10; //if the sequence is shorter than the barcode then bail out + break; + } + if(rawRSequence.length() < roligo.length()){ //let's just assume that the barcodes are the same length + success = pdiffs + 10; //if the sequence is shorter than the barcode then bail out + break; + } + + if((compareDNASeq(foligo, rawFSequence.substr(0,foligo.length()))) && (compareDNASeq(roligo, rawRSequence.substr((rawRSequence.length()-roligo.length()),roligo.length())))) { + group = it->first; + forwardSeq.setUnaligned(rawFSequence.substr(foligo.length())); + reverseSeq.setUnaligned(rawRSequence.substr(0,(rawRSequence.length()-roligo.length()))); + success = 0; + 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 + Alignment* alignment; + if (ifprimers.size() > 0) { alignment = new NeedlemanOverlap(-1.0, 1.0, -1.0, (maxFPrimerLength+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=rbarcodes.begin();it!=rbarcodes.end();it++){ + vector< vector > minFGroup; + vector minFPos; + + //the pair can have duplicates, but we only want to search for the unique forward and reverses, example + /* + 1 Sarah Westcott + 2 John Westcott + 3 Anna Westcott + 4 Sarah Schloss + 5 Pat Schloss + 6 Gail Brown + 7 Pat Moore + + only want to look for forward = Sarah, John, Anna, Pat, Gail + reverse = Westcott, Schloss, Brown, Moore + + but if best match forward = 4, and reverse = 1, we want to count as a valid match because forward 1 and forward 4 are the same. so both barcodes map to same group. + */ + //cout << endl << forwardSeq.getName() << endl; + for(map >::iterator it=ifprimers.begin();it!=ifprimers.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; + if(rawFSequence.length() < maxFPrimerLength){ //let's just assume that the barcodes are the same length + success = pdiffs + 10; break; } - + //cout << "before = " << oligo << '\t' << rawFSequence.substr(0,oligo.length()+pdiffs) << endl; //use needleman to align first barcode.length()+numdiffs of sequence to each barcode - alignment->align(oligo, rawSequence.substr((rawSequence.length()-(oligo.length()+bdiffs)),oligo.length()+bdiffs)); + alignment->align(oligo, rawFSequence.substr(0,oligo.length()+pdiffs)); oligo = alignment->getSeqAAln(); string temp = alignment->getSeqBAln(); - + int alnLength = oligo.length(); - for(int i=0;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 (alnLength == 0) { numDiff = pdiffs + 100; } + //cout << "after = " << oligo << '\t' << temp << '\t' << numDiff << endl; + if(numDiff < minDiff){ minDiff = numDiff; minCount = 1; - minGroup = it->second; - minPos = 0; - for(int i=alnLength-1;i>=0;i--){ + minFGroup.clear(); + minFGroup.push_back(it->second); + int tempminFPos = 0; + minFPos.clear(); + for(int i=0;isecond); + int tempminFPos = 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(0, (rawSequence.length()-minPos))); + + //cout << minDiff << '\t' << minCount << '\t' << endl; + if(minDiff > pdiffs) { success = minDiff; } //no good matches + else{ + //check for reverse match + if (alignment != NULL) { delete alignment; } - if(qual.getName() != ""){ - qual.trimQScores(-1, (rawSequence.length()-minPos)); - } - success = minDiff; + if (irbarcodes.size() > 0) { alignment = new NeedlemanOverlap(-1.0, 1.0, -1.0, (maxRPrimerLength+pdiffs+1)); } + else{ alignment = NULL; } + + //can you find the barcode + minDiff = 1e6; + minCount = 1; + vector< vector > minRGroup; + vector minRPos; + + for(map >::iterator it=irprimers.begin();it!=irprimers.end();it++){ + string oligo = it->first; + //cout << "before = " << oligo << '\t' << rawRSequence.substr(0,oligo.length()+pdiffs) << endl; + if(rawRSequence.length() < maxRPrimerLength){ //let's just assume that the barcodes are the same length + success = pdiffs + 10; + break; + } + + //use needleman to align first barcode.length()+numdiffs of sequence to each barcode + alignment->align(oligo, rawRSequence.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 (alnLength == 0) { numDiff = pdiffs + 100; } + + //cout << "after = " << oligo << '\t' << temp << '\t' << numDiff << endl; + if(numDiff < minDiff){ + minDiff = numDiff; + minCount = 1; + minRGroup.clear(); + minRGroup.push_back(it->second); + int tempminRPos = 0; + minRPos.clear(); + for(int i=0;isecond); + } + + } + + /*cout << minDiff << '\t' << minCount << '\t' << endl; + for (int i = 0; i < minFGroup.size(); i++) { + cout << i << '\t'; + for (int j = 0; j < minFGroup[i].size(); j++) { cout << minFGroup[i][j] << " "; } + cout << endl; + } + cout << endl; + for (int i = 0; i < minRGroup.size(); i++) { + cout << i << '\t'; + for (int j = 0; j < minRGroup[i].size(); j++) { cout << minRGroup[i][j] << " "; } + cout << endl; + } + cout << endl;*/ + if(minDiff > pdiffs) { success = minDiff; } //no good matches + else { + bool foundMatch = false; + vector matches; + for (int i = 0; i < minFGroup.size(); i++) { + for (int j = 0; j < minFGroup[i].size(); j++) { + for (int k = 0; k < minRGroup.size(); k++) { + if (m->inUsersGroups(minFGroup[i][j], minRGroup[k])) { matches.push_back(minFGroup[i][j]); k+= minRGroup.size(); } + } + } + } + + int fStart = 0; + int rStart = 0; + if (matches.size() == 1) { + foundMatch = true; + group = matches[0]; + fStart = minFPos[0]; + rStart = minRPos[0]; + } + + //we have an acceptable match for the forward and reverse, but do they match? + if (foundMatch) { + forwardSeq.setUnaligned(rawFSequence.substr(fStart)); + reverseSeq.setUnaligned(rawRSequence.substr(rStart)); + success = minDiff; + }else { success = pdiffs + 100; } + } } if (alignment != NULL) { delete alignment; } - } return success; } catch(exception& e) { - m->errorOut(e, "TrimOligos", "stripRBarcode"); + m->errorOut(e, "TrimOligos", "stripIForward"); exit(1); } } //*******************************************************************/ -int TrimOligos::stripRBarcode(Sequence& seq, int& group){ +int TrimOligos::stripBarcode(Sequence& seq, int& group){ try { string rawSequence = seq.getUnaligned(); int success = bdiffs + 1; //guilty until proven innocent //can you find the barcode - for(map::iterator it=rbarcodes.begin();it!=rbarcodes.end();it++){ + 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((rawSequence.length()-oligo.length()),oligo.length()))){ + + if(compareDNASeq(oligo, rawSequence.substr(0,oligo.length()))){ group = it->second; - seq.setUnaligned(rawSequence.substr(0,(rawSequence.length()-oligo.length()))); + seq.setUnaligned(rawSequence.substr(oligo.length())); success = 0; break; @@ -456,21 +1340,9 @@ int TrimOligos::stripRBarcode(Sequence& seq, int& group){ if ((bdiffs == 0) || (success == 0)) { return success; } else { //try aligning and see if you can find it - - int maxLength = 0; - - Alignment* alignment; - if (rbarcodes.size() > 0) { - map::iterator it; - - for(it=rbarcodes.begin();it!=rbarcodes.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; } + Alignment* alignment; + if (barcodes.size() > 0) { alignment = new NeedlemanOverlap(-1.0, 1.0, -1.0, (maxFBarcodeLength+bdiffs+1)); } + else{ alignment = NULL; } //can you find the barcode int minDiff = 1e6; @@ -478,28 +1350,28 @@ int TrimOligos::stripRBarcode(Sequence& seq, int& group){ int minGroup = -1; int minPos = 0; - for(map::iterator it=rbarcodes.begin();it!=rbarcodes.end();it++){ + 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 + if(rawSequence.length() < maxFBarcodeLength){ //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((rawSequence.length()-(oligo.length()+bdiffs)),oligo.length()+bdiffs)); + alignment->align(oligo, rawSequence.substr(0,oligo.length()+bdiffs)); oligo = alignment->getSeqAAln(); string temp = alignment->getSeqBAln(); - + int alnLength = oligo.length(); - for(int i=0;i=0;i--){ + if(oligo[i] != '-'){ alnLength = i+1; break; } } - oligo = oligo.substr(alnLength); - temp = temp.substr(alnLength); - + oligo = oligo.substr(0,alnLength); + temp = temp.substr(0,alnLength); + int numDiff = countDiffs(oligo, temp); if(numDiff < minDiff){ @@ -507,7 +1379,7 @@ int TrimOligos::stripRBarcode(Sequence& seq, int& group){ minCount = 1; minGroup = it->second; minPos = 0; - for(int i=alnLength-1;i>=0;i--){ + for(int i=0;i 1) { success = bdiffs + 100; } //can't tell the difference between multiple barcodes else{ //use the best match group = minGroup; - seq.setUnaligned(rawSequence.substr(0, (rawSequence.length()-minPos))); - + seq.setUnaligned(rawSequence.substr(minPos)); success = minDiff; } @@ -536,12 +1407,13 @@ int TrimOligos::stripRBarcode(Sequence& seq, int& group){ } catch(exception& e) { - m->errorOut(e, "TrimOligos", "stripRBarcode"); + m->errorOut(e, "TrimOligos", "stripBarcode"); exit(1); } } -//********************************************************************/ + +/********************************************************************/ int TrimOligos::stripForward(Sequence& seq, int& group){ try { @@ -568,21 +1440,9 @@ int TrimOligos::stripForward(Sequence& seq, int& group){ 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; - - for(it=primers.begin();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; } + if (primers.size() > 0) { alignment = new NeedlemanOverlap(-1.0, 1.0, -1.0, (maxFPrimerLength+pdiffs+1)); } + else{ alignment = NULL; } //can you find the barcode int minDiff = 1e6; @@ -594,7 +1454,7 @@ int TrimOligos::stripForward(Sequence& seq, int& group){ string oligo = it->first; // int length = oligo.length(); - if(rawSequence.length() < maxLength){ + if(rawSequence.length() < maxFPrimerLength){ success = pdiffs + 100; break; } @@ -680,21 +1540,9 @@ int TrimOligos::stripForward(Sequence& seq, QualityScores& qual, int& group, boo 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; - - for(it=primers.begin();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; } + if (primers.size() > 0) { alignment = new NeedlemanOverlap(-1.0, 1.0, -1.0, (maxFPrimerLength+pdiffs+1)); } + else{ alignment = NULL; } //can you find the barcode int minDiff = 1e6; @@ -706,7 +1554,7 @@ int TrimOligos::stripForward(Sequence& seq, QualityScores& qual, int& group, boo string oligo = it->first; // int length = oligo.length(); - if(rawSequence.length() < maxLength){ + if(rawSequence.length() < maxFPrimerLength){ success = pdiffs + 100; break; } @@ -855,19 +1703,9 @@ bool TrimOligos::stripLinker(Sequence& seq, QualityScores& qual){ if ((ldiffs == 0) || (success == 0)) { return success; } else { //try aligning and see if you can find it - - int maxLength = 0; - Alignment* alignment; - if (linker.size() > 0) { - for(int i = 0; i < linker.size(); i++){ - if(linker[i].length() > maxLength){ - maxLength = linker[i].length(); - } - } - alignment = new NeedlemanOverlap(-1.0, 1.0, -1.0, (maxLength+ldiffs+1)); - - }else{ alignment = NULL; } + if (linker.size() > 0) { alignment = new NeedlemanOverlap(-1.0, 1.0, -1.0, (maxLinkerLength+ldiffs+1)); } + else{ alignment = NULL; } //can you find the barcode int minDiff = 1e6; @@ -878,7 +1716,7 @@ bool TrimOligos::stripLinker(Sequence& seq, QualityScores& qual){ string oligo = linker[i]; // int length = oligo.length(); - if(rawSequence.length() < maxLength){ //let's just assume that the barcodes are the same length + if(rawSequence.length() < maxLinkerLength){ //let's just assume that the barcodes are the same length success = ldiffs + 10; break; } @@ -964,19 +1802,9 @@ bool TrimOligos::stripLinker(Sequence& seq){ if ((ldiffs == 0) || (success == 0)) { return success; } else { //try aligning and see if you can find it - - int maxLength = 0; - Alignment* alignment; - if (linker.size() > 0) { - for(int i = 0; i < linker.size(); i++){ - if(linker[i].length() > maxLength){ - maxLength = linker[i].length(); - } - } - alignment = new NeedlemanOverlap(-1.0, 1.0, -1.0, (maxLength+ldiffs+1)); - - }else{ alignment = NULL; } + if (linker.size() > 0) {alignment = new NeedlemanOverlap(-1.0, 1.0, -1.0, (maxLinkerLength+ldiffs+1)); } + else{ alignment = NULL; } //can you find the barcode int minDiff = 1e6; @@ -987,7 +1815,7 @@ bool TrimOligos::stripLinker(Sequence& seq){ string oligo = linker[i]; // int length = oligo.length(); - if(rawSequence.length() < maxLength){ //let's just assume that the barcodes are the same length + if(rawSequence.length() < maxLinkerLength){ //let's just assume that the barcodes are the same length success = ldiffs + 10; break; } @@ -1071,19 +1899,9 @@ bool TrimOligos::stripSpacer(Sequence& seq, QualityScores& qual){ if ((sdiffs == 0) || (success == 0)) { return success; } else { //try aligning and see if you can find it - - int maxLength = 0; - Alignment* alignment; - if (spacer.size() > 0) { - for(int i = 0; i < spacer.size(); i++){ - if(spacer[i].length() > maxLength){ - maxLength = spacer[i].length(); - } - } - alignment = new NeedlemanOverlap(-1.0, 1.0, -1.0, (maxLength+sdiffs+1)); - - }else{ alignment = NULL; } + if (spacer.size() > 0) { alignment = new NeedlemanOverlap(-1.0, 1.0, -1.0, (maxSpacerLength+sdiffs+1)); } + else{ alignment = NULL; } //can you find the barcode int minDiff = 1e6; @@ -1094,7 +1912,7 @@ bool TrimOligos::stripSpacer(Sequence& seq, QualityScores& qual){ string oligo = spacer[i]; // int length = oligo.length(); - if(rawSequence.length() < maxLength){ //let's just assume that the barcodes are the same length + if(rawSequence.length() < maxSpacerLength){ //let's just assume that the barcodes are the same length success = sdiffs + 10; break; } @@ -1180,19 +1998,9 @@ bool TrimOligos::stripSpacer(Sequence& seq){ if ((sdiffs == 0) || (success == 0)) { return success; } else { //try aligning and see if you can find it - - int maxLength = 0; - Alignment* alignment; - if (spacer.size() > 0) { - for(int i = 0; i < spacer.size(); i++){ - if(spacer[i].length() > maxLength){ - maxLength = spacer[i].length(); - } - } - alignment = new NeedlemanOverlap(-1.0, 1.0, -1.0, (maxLength+sdiffs+1)); - - }else{ alignment = NULL; } + if (spacer.size() > 0) { alignment = new NeedlemanOverlap(-1.0, 1.0, -1.0, (maxSpacerLength+sdiffs+1)); } + else{ alignment = NULL; } //can you find the barcode int minDiff = 1e6; @@ -1203,7 +2011,7 @@ bool TrimOligos::stripSpacer(Sequence& seq){ string oligo = spacer[i]; // int length = oligo.length(); - if(rawSequence.length() < maxLength){ //let's just assume that the barcodes are the same length + if(rawSequence.length() < maxSpacerLength){ //let's just assume that the barcodes are the same length success = sdiffs + 10; break; }