FastqRead::FastqRead(Sequence s, QualityScores q, string f) {
try {
m = MothurOut::getInstance(); format = f;
//fill convert table - goes from solexa to sanger. Used fq_all2std.pl as a reference.
for (int i = -64; i < 65; i++) {
char temp = (char) ((int)(33 + 10*log(1+pow(10,(i/10.0)))/log(10)+0.499));
convertTable.push_back(temp);
convertBackTable.push_back(((int)(33 + 10*log(1+pow(10,(i/10.0)))/log(10)+0.499)));
}
if (s.getName() != q.getName()) { m->mothurOut("[ERROR]: sequence name does not match quality score name. Cannot construct fastq object.\n"); m->control_pressed = true; }
else {
name = s.getName();
comment = s.getComment();
sequence = s.getUnaligned();
scores = q.getScores();
scoreString = convertQual(scores);
}
}
catch(exception& e) {
m->errorOut(e, "FastqRead", "FastqRead");
exit(1);
}
}
/********************************************************************/
TrimOligos::~TrimOligos() {}
//*******************************************************************/
int TrimOligos::stripBarcode(Sequence& seq, QualityScores& qual, int& group){
try {
string rawSequence = seq.getUnaligned();
int success = bdiffs + 1; //guilty until proven innocent
//can you find the barcode
for(map<string,int>::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;
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<string,int>::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<string,int>::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<alnLength;i++){
if(temp[i] != '-'){
minPos++;
}
}
}
else if(numDiff == minDiff){
minCount++;
}
}
if(minDiff > 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, "TrimOligos", "stripBarcode");
exit(1);
}
}
//*******************************************************************/
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<string,int>::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()));
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
int maxLength = 0;
Alignment* alignment;
if (barcodes.size() > 0) {
map<string,int>::iterator it=barcodes.begin();
for(map<string,int>::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; }
//can you find the barcode
int minDiff = 1e6;
int minCount = 1;
int minGroup = -1;
int minPos = 0;
for(map<string,int>::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<alnLength;i++){
if(temp[i] != '-'){
minPos++;
}
}
}
else if(numDiff == minDiff){
minCount++;
}
}
if(minDiff > 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;
}
if (alignment != NULL) { delete alignment; }
}
return success;
}
catch(exception& e) {
m->errorOut(e, "TrimOligos", "stripBarcode");
exit(1);
}
}
//********************************************************************/
int TrimOligos::stripForward(Sequence& seq, int& group){
try {
string rawSequence = seq.getUnaligned();
int success = pdiffs + 1; //guilty until proven innocent
//can you find the primer
for(map<string,int>::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
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<string,int>::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<string,int>::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<alnLength;i++){
if(temp[i] != '-'){
minPos++;
}
}
}
else if(numDiff == minDiff){
minCount++;
}
}
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;
}
if (alignment != NULL) { delete alignment; }
}
return success;
}
catch(exception& e) {
m->errorOut(e, "TrimOligos", "stripForward");
exit(1);
}
}
//*******************************************************************/
int TrimOligos::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<string,int>::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;
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<string,int>::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<string,int>::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<alnLength;i++){
if(temp[i] != '-'){
minPos++;
}
}
}
else if(numDiff == minDiff){
minCount++;
}
}
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));
if(qual.getName() != ""){
qual.trimQScores(minPos, -1);
}
success = minDiff;
}
if (alignment != NULL) { delete alignment; }
}
return success;
}
catch(exception& e) {
m->errorOut(e, "TrimOligos", "stripForward");
exit(1);
}
}
//******************************************************************/
bool TrimOligos::stripReverse(Sequence& seq, QualityScores& qual){
try {
string rawSequence = seq.getUnaligned();
bool success = 0; //guilty until proven innocent
for(int i=0;i<revPrimer.size();i++){
string oligo = revPrimer[i];
if(rawSequence.length() < oligo.length()){
success = 0;
break;
}
if(compareDNASeq(oligo, rawSequence.substr(rawSequence.length()-oligo.length(),oligo.length()))){
seq.setUnaligned(rawSequence.substr(0,rawSequence.length()-oligo.length()));
if(qual.getName() != ""){
qual.trimQScores(-1, rawSequence.length()-oligo.length());
}
success = 1;
break;
}
}
return success;
}
catch(exception& e) {
m->errorOut(e, "TrimOligos", "stripReverse");
exit(1);
}
}
int ReverseSeqsCommand::execute(){
try{
if (abort == true) { if (calledHelp) { return 0; } return 2; }
string fastaReverseFileName;
if(fastaFileName != ""){
ifstream inFASTA;
m->openInputFile(fastaFileName, inFASTA);
ofstream outFASTA;
string tempOutputDir = outputDir;
if (outputDir == "") { tempOutputDir += m->hasPath(fastaFileName); } //if user entered a file with a path then preserve it
map<string, string> variables;
variables["[filename]"] = tempOutputDir + m->getRootName(m->getSimpleName(fastaFileName));
variables["[extension]"] = m->getExtension(fastaFileName);
fastaReverseFileName = getOutputFileName("fasta", variables);
m->openOutputFile(fastaReverseFileName, outFASTA);
while(!inFASTA.eof()){
if (m->control_pressed) { inFASTA.close(); outFASTA.close(); m->mothurRemove(fastaReverseFileName); return 0; }
Sequence currSeq(inFASTA); m->gobble(inFASTA);
if (currSeq.getName() != "") {
currSeq.reverseComplement();
currSeq.printSequence(outFASTA);
}
}
inFASTA.close();
outFASTA.close();
outputNames.push_back(fastaReverseFileName); outputTypes["fasta"].push_back(fastaReverseFileName);
}
string qualReverseFileName;
if(qualFileName != ""){
QualityScores currQual;
ifstream inQual;
m->openInputFile(qualFileName, inQual);
ofstream outQual;
string tempOutputDir = outputDir;
if (outputDir == "") { tempOutputDir += m->hasPath(qualFileName); } //if user entered a file with a path then preserve it
map<string, string> variables;
variables["[filename]"] = tempOutputDir + m->getRootName(m->getSimpleName(qualFileName));
variables["[extension]"] = m->getExtension(qualFileName);
string qualReverseFileName = getOutputFileName("qfile", variables);
m->openOutputFile(qualReverseFileName, outQual);
while(!inQual.eof()){
if (m->control_pressed) { inQual.close(); outQual.close(); m->mothurRemove(qualReverseFileName); return 0; }
currQual = QualityScores(inQual); m->gobble(inQual);
currQual.flipQScores();
currQual.printQScores(outQual);
}
inQual.close();
outQual.close();
outputNames.push_back(qualReverseFileName); outputTypes["qfile"].push_back(qualReverseFileName);
}
if (m->control_pressed) { m->mothurRemove(qualReverseFileName); m->mothurRemove(fastaReverseFileName); return 0; }
//set fasta file as new current fastafile
string current = "";
itTypes = outputTypes.find("fasta");
if (itTypes != outputTypes.end()) {
if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setFastaFile(current); }
}
itTypes = outputTypes.find("qfile");
if (itTypes != outputTypes.end()) {
if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setQualFile(current); }
}
m->mothurOutEndLine();
m->mothurOut("Output File Names: "); m->mothurOutEndLine();
for(int i=0;i<outputNames.size();i++){
m->mothurOut(outputNames[i]);
m->mothurOutEndLine();
}
return 0;
}
catch(exception& e) {
m->errorOut(e, "ReverseSeqsCommand", "execute");
exit(1);
}
}
/********************************************************************/
TrimOligos::~TrimOligos() {}
//*******************************************************************/
int TrimOligos::stripBarcode(Sequence& seq, QualityScores& qual, int& group){
try {
string rawSequence = seq.getUnaligned();
int success = bdiffs + 1; //guilty until proven innocent
//can you find the barcode
for(map<string,int>::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;
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<string,int>::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<string,int>::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<alnLength;i++){
if(temp[i] != '-'){
minPos++;
}
}
}
else if(numDiff == minDiff){
minCount++;
}
}
if(minDiff > 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, "TrimOligos", "stripBarcode");
exit(1);
}
}
//*******************************************************************/
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<string,int>::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()));
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
int maxLength = 0;
Alignment* alignment;
if (barcodes.size() > 0) {
map<string,int>::iterator it=barcodes.begin();
for(map<string,int>::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; }
//can you find the barcode
int minDiff = 1e6;
int minCount = 1;
int minGroup = -1;
int minPos = 0;
for(map<string,int>::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<alnLength;i++){
if(temp[i] != '-'){
minPos++;
}
}
}
else if(numDiff == minDiff){
minCount++;
}
}
if(minDiff > 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;
}
if (alignment != NULL) { delete alignment; }
}
return success;
}
catch(exception& e) {
m->errorOut(e, "TrimOligos", "stripBarcode");
exit(1);
}
}
//********************************************************************/
int TrimOligos::stripForward(Sequence& seq, int& group){
try {
string rawSequence = seq.getUnaligned();
int success = pdiffs + 1; //guilty until proven innocent
//can you find the primer
for(map<string,int>::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
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<string,int>::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<string,int>::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<alnLength;i++){
if(temp[i] != '-'){
minPos++;
}
}
}
else if(numDiff == minDiff){
minCount++;
}
}
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;
}
if (alignment != NULL) { delete alignment; }
}
return success;
}
catch(exception& e) {
m->errorOut(e, "TrimOligos", "stripForward");
exit(1);
}
}
//*******************************************************************/
int TrimOligos::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<string,int>::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;
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<string,int>::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<string,int>::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<alnLength;i++){
if(temp[i] != '-'){
minPos++;
}
}
}
else if(numDiff == minDiff){
minCount++;
}
}
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));
if(qual.getName() != ""){
qual.trimQScores(minPos, -1);
}
success = minDiff;
}
if (alignment != NULL) { delete alignment; }
}
return success;
}
catch(exception& e) {
m->errorOut(e, "TrimOligos", "stripForward");
exit(1);
}
}
