#include "trimseqscommand.h"
#include "needlemanoverlap.hpp"
+#include "trimoligos.h"
//**********************************************************************************************************************
vector<string> 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);
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";
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";
// ...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 = ""; }
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 { 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);
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 == "")){
m->mothurOut("You didn't set any options... quiting command."); m->mothurOutEndLine();
abort = true;
}
+
+ if (nameFile == "") {
+ vector<string> files; files.push_back(fastaFile);
+ parser.getNameFile(files);
+ }
}
}
}
}
- vector<unsigned long int> fastaFilePos;
- vector<unsigned long int> qFilePos;
+ //fills lines and qlines
+ setLines(fastaFile, qFileName);
- setLines(fastaFile, qFileName, fastaFilePos, 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);
+ }
- 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; }
/**************************************************************************************/
-int TrimSeqsCommand::driverCreateTrim(string filename, string qFileName, string trimFileName, string scrapFileName, string trimQFileName, string scrapQFileName, string trimNFileName, string scrapNFileName, string groupFileName, vector<vector<string> > fastaFileNames, vector<vector<string> > qualFileNames, vector<vector<string> > 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<vector<string> > fastaFileNames, vector<vector<string> > qualFileNames, vector<vector<string> > nameFileNames, linePair line, linePair qline) {
try {
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, revPrimer, linker, spacer);
while (moreSeqs) {
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(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; }
}
if (currentSeqsDiffs > tdiffs) { trashCode += 't'; }
if(numRPrimers != 0){
- success = stripReverse(currSeq, currQual);
+ success = trimOligos.stripReverse(currSeq, currQual);
if(!success) { trashCode += 'r'; }
}
}
#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
- unsigned long int pos = inFASTA.tellg();
- if ((pos == -1) || (pos >= line->end)) { break; }
+ unsigned long long pos = inFASTA.tellg();
+ if ((pos == -1) || (pos >= line.end)) { break; }
#else
if (inFASTA.eof()) { break; }
int TrimSeqsCommand::createProcessesCreateTrim(string filename, string qFileName, string trimFASTAFileName, string scrapFASTAFileName, string trimQualFileName, string scrapQualFileName, string trimNameFileName, string scrapNameFileName, string groupFile, vector<vector<string> > fastaFileNames, vector<vector<string> > qualFileNames, vector<vector<string> > 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)
+ //loop through and create all the processes you want
while (process != processors) {
int pid = fork();
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<trimData*> pDataArray;
+ DWORD dwThreadIdArray[processors-1];
+ HANDLE hThreadArray[processors-1];
+
+ //Create processor worker threads.
+ for( int i=0; i<processors-1; i++){
+
+ string extension = "";
+ if (i != 0) { extension = toString(i) + ".temp"; processIDS.push_back(i); }
+ vector<vector<string> > tempFASTAFileNames = fastaFileNames;
+ vector<vector<string> > tempPrimerQualFileNames = qualFileNames;
+ vector<vector<string> > tempNameFileNames = nameFileNames;
+
+ if(allFiles){
+ ofstream temp;
+
+ for(int i=0;i<tempFASTAFileNames.size();i++){
+ for(int j=0;j<tempFASTAFileNames[i].size();j++){
+ if (tempFASTAFileNames[i][j] != "") {
+ tempFASTAFileNames[i][j] += extension;
+ m->openOutputFile(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, 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<string, int>::iterator it = pDataArray[i]->groupCounts.begin(); it != pDataArray[i]->groupCounts.end(); it++) {
+ map<string, int>::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;i<processIDS.size();i++){
m->mothurOut("Appending files from process " + toString(processIDS[i])); m->mothurOutEndLine();
}
}
+ #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
if(createGroup){
ifstream in;
string tempFile = filename + toString(processIDS[i]) + ".num.temp";
if (tempNum != 0) {
while (!in.eof()) {
in >> group >> tempNum; m->gobble(in);
-
+
map<string, int>::iterator it = groupCounts.find(group);
if (it == groupCounts.end()) { groupCounts[group] = tempNum; }
else { groupCounts[it->first] += tempNum; }
}
in.close(); m->mothurRemove(tempFile);
}
-
+ #endif
}
-
- return exitCommand;
-#endif
+
+ return exitCommand;
}
catch(exception& e) {
m->errorOut(e, "TrimSeqsCommand", "createProcessesCreateTrim");
/**************************************************************************************************/
-int TrimSeqsCommand::setLines(string filename, string qfilename, vector<unsigned long int>& fastaFilePos, vector<unsigned long int>& qfileFilePos) {
+int TrimSeqsCommand::setLines(string filename, string qfilename) {
try {
+
+ vector<unsigned long long> fastaFilePos;
+ vector<unsigned long long> qfileFilePos;
+
#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
//set file positions for fasta file
fastaFilePos = m->divideFile(filename, processors);
map<string, int>::iterator it = firstSeqNames.find(sname);
if(it != firstSeqNames.end()) { //this is the start of a new chunk
- unsigned long int pos = inQual.tellg();
+ unsigned long long pos = inQual.tellg();
qfileFilePos.push_back(pos - input.length() - 1);
firstSeqNames.erase(it);
}
//get last file position of qfile
FILE * pFile;
- unsigned long int size;
+ unsigned long long size;
//get num bytes in file
pFile = fopen (qfilename.c_str(),"rb");
}
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 (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));
+ cout << fastaFilePos[startIndex] << '\t' << numSeqsPerProcessor << endl;
+ if (qfilename != "") { qLines.push_back(linePair(qfileFilePos[startIndex], numSeqsPerProcessor)); }
+ }
+
+ if(qfilename == "") { qLines = lines; } //files with duds
+ }
return 1;
#endif
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(); }
}
}
numFPrimers = primers.size();
numRPrimers = revPrimer.size();
+ numLinkers = linker.size();
+ numSpacers = spacer.size();
bool allBlank = true;
for (int i = 0; i < barcodeNameVector.size(); i++) {
exit(1);
}
}
-
-//***************************************************************************************************************
-
-int TrimSeqsCommand::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, "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<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, "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;i<numRPrimers;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, "TrimSeqsCommand", "stripReverse");
- exit(1);
- }
-}
-
//***************************************************************************************************************
bool TrimSeqsCommand::keepFirstTrim(Sequence& sequence, QualityScores& qscores){
}
}
-
-//***************************************************************************************************************
-
-bool TrimSeqsCommand::compareDNASeq(string oligo, string seq){
- try {
- bool success = 1;
- int length = oligo.length();
-
- for(int i=0;i<length;i++){
-
- if(oligo[i] != seq[i]){
- if(oligo[i] == 'A' || oligo[i] == 'T' || oligo[i] == 'G' || oligo[i] == 'C') { success = 0; }
- else if((oligo[i] == 'N' || oligo[i] == 'I') && (seq[i] == 'N')) { success = 0; }
- else if(oligo[i] == 'R' && (seq[i] != 'A' && seq[i] != 'G')) { success = 0; }
- else if(oligo[i] == 'Y' && (seq[i] != 'C' && seq[i] != 'T')) { success = 0; }
- else if(oligo[i] == 'M' && (seq[i] != 'C' && seq[i] != 'A')) { success = 0; }
- else if(oligo[i] == 'K' && (seq[i] != 'T' && seq[i] != 'G')) { success = 0; }
- else if(oligo[i] == 'W' && (seq[i] != 'T' && seq[i] != 'A')) { success = 0; }
- else if(oligo[i] == 'S' && (seq[i] != 'C' && seq[i] != 'G')) { success = 0; }
- else if(oligo[i] == 'B' && (seq[i] != 'C' && seq[i] != 'T' && seq[i] != 'G')) { success = 0; }
- else if(oligo[i] == 'D' && (seq[i] != 'A' && seq[i] != 'T' && seq[i] != 'G')) { success = 0; }
- else if(oligo[i] == 'H' && (seq[i] != 'A' && seq[i] != 'T' && seq[i] != 'C')) { success = 0; }
- else if(oligo[i] == 'V' && (seq[i] != 'A' && seq[i] != 'C' && seq[i] != 'G')) { success = 0; }
-
- if(success == 0) { break; }
- }
- else{
- success = 1;
- }
- }
-
- return success;
- }
- catch(exception& e) {
- m->errorOut(e, "TrimSeqsCommand", "compareDNASeq");
- exit(1);
- }
-
-}
-
-//***************************************************************************************************************
-
-int TrimSeqsCommand::countDiffs(string oligo, string seq){
- try {
-
- int length = oligo.length();
- int countDiffs = 0;
-
- for(int i=0;i<length;i++){
-
- if(oligo[i] != seq[i]){
- if(oligo[i] == 'A' || oligo[i] == 'T' || oligo[i] == 'G' || oligo[i] == 'C' || oligo[i] == '-' || oligo[i] == '.') { countDiffs++; }
- else if((oligo[i] == 'N' || oligo[i] == 'I') && (seq[i] == 'N')) { countDiffs++; }
- else if(oligo[i] == 'R' && (seq[i] != 'A' && seq[i] != 'G')) { countDiffs++; }
- else if(oligo[i] == 'Y' && (seq[i] != 'C' && seq[i] != 'T')) { countDiffs++; }
- else if(oligo[i] == 'M' && (seq[i] != 'C' && seq[i] != 'A')) { countDiffs++; }
- else if(oligo[i] == 'K' && (seq[i] != 'T' && seq[i] != 'G')) { countDiffs++; }
- else if(oligo[i] == 'W' && (seq[i] != 'T' && seq[i] != 'A')) { countDiffs++; }
- else if(oligo[i] == 'S' && (seq[i] != 'C' && seq[i] != 'G')) { countDiffs++; }
- else if(oligo[i] == 'B' && (seq[i] != 'C' && seq[i] != 'T' && seq[i] != 'G')) { countDiffs++; }
- else if(oligo[i] == 'D' && (seq[i] != 'A' && seq[i] != 'T' && seq[i] != 'G')) { countDiffs++; }
- else if(oligo[i] == 'H' && (seq[i] != 'A' && seq[i] != 'T' && seq[i] != 'C')) { countDiffs++; }
- else if(oligo[i] == 'V' && (seq[i] != 'A' && seq[i] != 'C' && seq[i] != 'G')) { countDiffs++; }
- }
-
- }
-
- return countDiffs;
- }
- catch(exception& e) {
- m->errorOut(e, "TrimSeqsCommand", "countDiffs");
- exit(1);
- }
-
-}
-
//***************************************************************************************************************