*/
#include "trimseqscommand.h"
+#include "needlemanoverlap.hpp"
//***************************************************************************************************************
else {
//valid paramters for this command
string AlignArray[] = {"fasta", "flip", "oligos", "maxambig", "maxhomop", "minlength", "maxlength", "qfile",
- "qthreshold", "qaverage", "allfiles", "qtrim", "outputdir","inputdir"};
+ "qthreshold", "qaverage", "allfiles", "qtrim","tdiffs", "pdiffs", "bdiffs", "processors", "outputdir","inputdir"};
vector<string> myArray (AlignArray, AlignArray+(sizeof(AlignArray)/sizeof(string)));
temp = validParameter.validFile(parameters, "maxlength", false); if (temp == "not found") { temp = "0"; }
convert(temp, maxLength);
+
+ temp = validParameter.validFile(parameters, "bdiffs", false); if (temp == "not found") { temp = "0"; }
+ convert(temp, bdiffs);
+
+ temp = validParameter.validFile(parameters, "pdiffs", false); if (temp == "not found") { temp = "0"; }
+ convert(temp, pdiffs);
+
+ temp = validParameter.validFile(parameters, "tdiffs", false); if (temp == "not found") { int tempTotal = pdiffs + bdiffs; temp = toString(tempTotal); }
+ convert(temp, tdiffs);
+
+ if(tdiffs == 0){ tdiffs = bdiffs + pdiffs; }
+
temp = validParameter.validFile(parameters, "qfile", true);
if (temp == "not found") { qFileName = ""; }
else if(temp == "not open") { abort = true; }
temp = validParameter.validFile(parameters, "allfiles", false); if (temp == "not found") { temp = "F"; }
allFiles = isTrue(temp);
+ temp = validParameter.validFile(parameters, "processors", false); if (temp == "not found"){ temp = "1"; }
+ convert(temp, processors);
+
if(allFiles && oligoFile == ""){
m->mothurOut("You selected allfiles, but didn't enter an oligos file. Ignoring the allfiles request."); m->mothurOutEndLine();
}
void TrimSeqsCommand::help(){
try {
m->mothurOut("The trim.seqs command reads a fastaFile and creates .....\n");
- m->mothurOut("The trim.seqs command parameters are fasta, flip, oligos, maxambig, maxhomop, minlength, maxlength, qfile, qthreshold, qaverage, qtrim and allfiles.\n");
+ m->mothurOut("The trim.seqs command parameters are fasta, flip, oligos, maxambig, maxhomop, minlength, maxlength, qfile, qthreshold, qaverage, diffs, qtrim and allfiles.\n");
m->mothurOut("The fasta parameter is required.\n");
- m->mothurOut("The flip parameter .... The default is 0.\n");
+ m->mothurOut("The flip parameter will output the reverse compliment of your trimmed sequence. The default is false.\n");
m->mothurOut("The oligos parameter .... The default is "".\n");
m->mothurOut("The maxambig parameter .... The default is -1.\n");
m->mothurOut("The maxhomop parameter .... The default is 0.\n");
m->mothurOut("The minlength parameter .... The default is 0.\n");
m->mothurOut("The maxlength parameter .... The default is 0.\n");
+ m->mothurOut("The tdiffs parameter is used to specify the total number of differences allowed in the sequence. The default is pdiffs + bdiffs.\n");
+ m->mothurOut("The bdiffs parameter is used to specify the number of differences allowed in the barcode. The default is 0.\n");
+ m->mothurOut("The pdiffs parameter is used to specify the number of differences allowed in the primer. The default is 0.\n");
m->mothurOut("The qfile parameter .....\n");
m->mothurOut("The qthreshold parameter .... The default is 0.\n");
m->mothurOut("The qaverage parameter .... The default is 0.\n");
numFPrimers = 0; //this needs to be initialized
numRPrimers = 0;
- ifstream inFASTA;
- openInputFile(fastaFile, inFASTA);
-
- ofstream outFASTA;
string trimSeqFile = outputDir + getRootName(getSimpleName(fastaFile)) + "trim.fasta";
- openOutputFile(trimSeqFile, outFASTA);
outputNames.push_back(trimSeqFile);
+ string scrapSeqFile = outputDir + getRootName(getSimpleName(fastaFile)) + "scrap.fasta";
+ outputNames.push_back(scrapSeqFile);
+ string groupFile = outputDir + getRootName(getSimpleName(fastaFile)) + "groups";
- ofstream outGroups;
- vector<ofstream*> fastaFileNames;
+ vector<string> fastaFileNames;
if(oligoFile != ""){
- string groupFile = outputDir + getRootName(getSimpleName(fastaFile)) + "groups";
- openOutputFile(groupFile, outGroups);
outputNames.push_back(groupFile);
getOligos(fastaFileNames);
}
+ if(qFileName != "") { setLines(qFileName, qLines); }
+
+
+ #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
+ if(processors == 1){
+ ifstream inFASTA;
+ int numSeqs;
+ openInputFile(fastaFile, inFASTA);
+ getNumSeqs(inFASTA, numSeqs);
+ inFASTA.close();
+
+ lines.push_back(new linePair(0, numSeqs));
+
+ driverCreateTrim(fastaFile, qFileName, trimSeqFile, scrapSeqFile, groupFile, fastaFileNames, lines[0], lines[0]);
+
+ for (int j = 0; j < fastaFileNames.size(); j++) {
+ rename((fastaFileNames[j] + toString(getpid()) + ".temp").c_str(), fastaFileNames[j].c_str());
+ }
+
+ }else{
+ setLines(fastaFile, lines);
+ if(qFileName == "") { qLines = lines; }
+
+ createProcessesCreateTrim(fastaFile, qFileName, trimSeqFile, scrapSeqFile, groupFile, fastaFileNames);
+
+ rename((trimSeqFile + toString(processIDS[0]) + ".temp").c_str(), trimSeqFile.c_str());
+ rename((scrapSeqFile + toString(processIDS[0]) + ".temp").c_str(), scrapSeqFile.c_str());
+ rename((groupFile + toString(processIDS[0]) + ".temp").c_str(), groupFile.c_str());
+ for (int j = 0; j < fastaFileNames.size(); j++) {
+ rename((fastaFileNames[j] + toString(processIDS[0]) + ".temp").c_str(), fastaFileNames[j].c_str());
+ }
+ //append files
+ for(int i=1;i<processors;i++){
+ appendFiles((trimSeqFile + toString(processIDS[i]) + ".temp"), trimSeqFile);
+ remove((trimSeqFile + toString(processIDS[i]) + ".temp").c_str());
+ appendFiles((scrapSeqFile + toString(processIDS[i]) + ".temp"), scrapSeqFile);
+ remove((scrapSeqFile + toString(processIDS[i]) + ".temp").c_str());
+ appendFiles((groupFile + toString(processIDS[i]) + ".temp"), groupFile);
+ remove((groupFile + toString(processIDS[i]) + ".temp").c_str());
+ for (int j = 0; j < fastaFileNames.size(); j++) {
+ appendFiles((fastaFileNames[j] + toString(processIDS[i]) + ".temp"), fastaFileNames[j]);
+ remove((fastaFileNames[j] + toString(processIDS[i]) + ".temp").c_str());
+ }
+ }
+ }
+
+ if (m->control_pressed) { return 0; }
+ #else
+ ifstream inFASTA;
+ int numSeqs;
+ openInputFile(fastaFile, inFASTA);
+ getNumSeqs(inFASTA, numSeqs);
+ inFASTA.close();
+
+ lines.push_back(new linePair(0, numSeqs));
+
+ driverCreateTrim(fastaFile, qFileName, trimSeqFile, scrapSeqFile, groupFile, fastaFileNames, lines[0], lines[0]);
+
+ if (m->control_pressed) { return 0; }
+ #endif
+
+
+ for(int i=0;i<fastaFileNames.size();i++){
+ ifstream inFASTA;
+ string seqName;
+ //openInputFile(getRootName(fastaFile) + groupVector[i] + ".fasta", inFASTA);
+ openInputFile(fastaFileNames[i], inFASTA);
+ ofstream outGroups;
+ string outGroupFilename = outputDir + getRootName(getSimpleName(fastaFileNames[i])) + "groups";
+ openOutputFile(outGroupFilename, outGroups);
+ //openOutputFile(outputDir + getRootName(getSimpleName(fastaFile)) + groupVector[i] + ".groups", outGroups);
+ outputNames.push_back(outGroupFilename);
+
+ while(!inFASTA.eof()){
+ if(inFASTA.get() == '>'){
+ inFASTA >> seqName;
+ outGroups << seqName << '\t' << groupVector[i] << endl;
+ }
+ while (!inFASTA.eof()) { char c = inFASTA.get(); if (c == 10 || c == 13){ break; } }
+ }
+ outGroups.close();
+ inFASTA.close();
+ }
+
+ if (m->control_pressed) {
+ for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); }
+ return 0;
+ }
+
+ m->mothurOutEndLine();
+ m->mothurOut("Output File Names: "); m->mothurOutEndLine();
+ for (int i = 0; i < outputNames.size(); i++) { m->mothurOut(outputNames[i]); m->mothurOutEndLine(); }
+ m->mothurOutEndLine();
+
+ return 0;
+
+ }
+ catch(exception& e) {
+ m->errorOut(e, "TrimSeqsCommand", "execute");
+ exit(1);
+ }
+}
+
+/**************************************************************************************/
+int TrimSeqsCommand::driverCreateTrim(string filename, string qFileName, string trimFile, string scrapFile, string groupFile, vector<string> fastaNames, linePair* line, linePair* qline) {
+ try {
+
+ ofstream outFASTA;
+ int able = openOutputFile(trimFile, outFASTA);
+
ofstream scrapFASTA;
- string scrapSeqFile = outputDir + getRootName(getSimpleName(fastaFile)) + "scrap.fasta";
- openOutputFile(scrapSeqFile, scrapFASTA);
- outputNames.push_back(scrapSeqFile);
+ openOutputFile(scrapFile, scrapFASTA);
+
+ ofstream outGroups;
+ vector<ofstream*> fastaFileNames;
+ if (oligoFile != "") {
+ openOutputFile(groupFile, outGroups);
+ for (int i = 0; i < fastaNames.size(); i++) {
+ #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
+ fastaFileNames.push_back(new ofstream((fastaNames[i] + toString(getpid()) + ".temp").c_str(), ios::ate));
+ #else
+ fastaFileNames.push_back(new ofstream((fastaNames[i] + toString(i) + ".temp").c_str(), ios::ate));
+ #endif
+ }
+ }
+
+ ifstream inFASTA;
+ openInputFile(filename, inFASTA);
ifstream qFile;
if(qFileName != "") { openInputFile(qFileName, qFile); }
- bool success;
-
- while(!inFASTA.eof()){
+ qFile.seekg(qline->start);
+ inFASTA.seekg(line->start);
+ for(int i=0;i<line->num;i++){
+
if (m->control_pressed) {
- inFASTA.close();
- outFASTA.close();
- scrapFASTA.close();
- outGroups.close();
- if(qFileName != "") { qFile.close(); }
- for(int i=0;i<fastaFileNames.size();i++){
- fastaFileNames[i]->close();
- delete fastaFileNames[i];
- }
+ inFASTA.close(); outFASTA.close(); scrapFASTA.close(); if (oligoFile != "") { outGroups.close(); } if(qFileName != "") { qFile.close(); }
+ for(int i=0;i<fastaFileNames.size();i++){ fastaFileNames[i]->close(); delete fastaFileNames[i]; }
for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); }
return 0;
}
-
+
+ int success = 1;
Sequence currSeq(inFASTA);
if (origSeq != "") {
int group;
string trashCode = "";
+ int currentSeqsDiffs = 0;
if(qFileName != ""){
if(qThreshold != 0) { success = stripQualThreshold(currSeq, qFile); }
else if(qAverage != 0) { success = cullQualAverage(currSeq, qFile); }
- if ((!qtrim) && (origSeq.length() != currSeq.getUnaligned().length())) {
+
+ if (qtrim == 1 && (origSeq.length() != currSeq.getUnaligned().length())) {
success = 0; //if you don't want to trim and the sequence does not meet quality requirements, move to scrap
}
+
if(!success) { trashCode += 'q'; }
}
if(barcodes.size() != 0){
success = stripBarcode(currSeq, group);
- if(!success){ trashCode += 'b'; }
+ if(success > bdiffs){ trashCode += 'b'; }
+ else{ currentSeqsDiffs += success; }
}
-
+
if(numFPrimers != 0){
success = stripForward(currSeq);
- if(!success){ trashCode += 'f'; }
+ if(success > pdiffs){ trashCode += 'f'; }
+ else{ currentSeqsDiffs += success; }
}
-
+
+ if (currentSeqsDiffs > tdiffs) { trashCode += 't'; }
+
if(numRPrimers != 0){
success = stripReverse(currSeq);
if(!success){ trashCode += 'r'; }
if(flip){ currSeq.reverseComplement(); } // should go last
if(trashCode.length() == 0){
- currSeq.setAligned(currSeq.getUnaligned()); //this is because of a modification we made to the sequence class to fix a bug. all seqs have an aligned version, which is the version that gets printed.
+ currSeq.setAligned(currSeq.getUnaligned());
currSeq.printSequence(outFASTA);
if(barcodes.size() != 0){
outGroups << currSeq.getName() << '\t' << groupVector[group] << endl;
else{
currSeq.setName(currSeq.getName() + '|' + trashCode);
currSeq.setUnaligned(origSeq);
+ currSeq.setAligned(origSeq);
currSeq.printSequence(scrapFASTA);
}
}
gobble(inFASTA);
}
+
inFASTA.close();
outFASTA.close();
scrapFASTA.close();
- outGroups.close();
+ if (oligoFile != "") { outGroups.close(); }
if(qFileName != "") { qFile.close(); }
for(int i=0;i<fastaFileNames.size();i++){
delete fastaFileNames[i];
}
- for(int i=0;i<fastaFileNames.size();i++){
- string seqName;
- openInputFile(getRootName(fastaFile) + groupVector[i] + ".fasta", inFASTA);
- ofstream outGroups;
- openOutputFile(outputDir + getRootName(getSimpleName(fastaFile)) + groupVector[i] + ".groups", outGroups);
- outputNames.push_back(outputDir + getRootName(getSimpleName(fastaFile)) + groupVector[i] + ".groups");
+ return 0;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "TrimSeqsCommand", "driverCreateTrim");
+ exit(1);
+ }
+}
+/**************************************************************************************************/
+int TrimSeqsCommand::createProcessesCreateTrim(string filename, string qFileName, string trimFile, string scrapFile, string groupFile, vector<string> fastaNames) {
+ try {
+#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
+ int process = 0;
+ int exitCommand = 1;
+ processIDS.clear();
+
+ //loop through and create all the processes you want
+ while (process != processors) {
+ int pid = fork();
- while(!inFASTA.eof()){
- if(inFASTA.get() == '>'){
- inFASTA >> seqName;
- outGroups << seqName << '\t' << groupVector[i] << endl;
- }
- while (!inFASTA.eof()) { char c = inFASTA.get(); if (c == 10 || c == 13){ break; } }
- }
- outGroups.close();
- inFASTA.close();
+ if (pid > 0) {
+ processIDS.push_back(pid); //create map from line number to pid so you can append files in correct order later
+ process++;
+ }else if (pid == 0){
+ driverCreateTrim(filename, qFileName, (trimFile + toString(getpid()) + ".temp"), (scrapFile + toString(getpid()) + ".temp"), (groupFile + toString(getpid()) + ".temp"), fastaNames, lines[process], qLines[process]);
+ exit(0);
+ }else { m->mothurOut("unable to spawn the necessary processes."); m->mothurOutEndLine(); exit(0); }
}
- if (m->control_pressed) {
- for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); }
- return 0;
+ //force parent to wait until all the processes are done
+ for (int i=0;i<processors;i++) {
+ int temp = processIDS[i];
+ wait(&temp);
}
+
+ return exitCommand;
+#endif
+ }
+ catch(exception& e) {
+ m->errorOut(e, "TrimSeqsCommand", "createProcessesCreateTrim");
+ exit(1);
+ }
+}
+/**************************************************************************************************/
- m->mothurOutEndLine();
- m->mothurOut("Output File Names: "); m->mothurOutEndLine();
- for (int i = 0; i < outputNames.size(); i++) { m->mothurOut(outputNames[i]); m->mothurOutEndLine(); }
- m->mothurOutEndLine();
+int TrimSeqsCommand::setLines(string filename, vector<linePair*>& lines) {
+ try {
+
+ lines.clear();
+
+ vector<long int> positions;
+
+ ifstream inFASTA;
+ openInputFile(filename, inFASTA);
- return 0;
+ string input;
+ while(!inFASTA.eof()){
+ input = getline(inFASTA);
+
+ if (input.length() != 0) {
+ if(input[0] == '>'){ long int pos = inFASTA.tellg(); positions.push_back(pos - input.length() - 1); }
+ }
+ }
+ inFASTA.close();
+
+ int numFastaSeqs = positions.size();
+
+ FILE * pFile;
+ long 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);
+ }
+
+ int numSeqsPerProcessor = numFastaSeqs / processors;
+
+ for (int i = 0; i < processors; i++) {
+
+ long int startPos = positions[ i * numSeqsPerProcessor ];
+ if(i == processors - 1){
+ numSeqsPerProcessor = numFastaSeqs - i * numSeqsPerProcessor;
+ }else{
+ long int myEnd = positions[ (i+1) * numSeqsPerProcessor ];
+ }
+ lines.push_back(new linePair(startPos, numSeqsPerProcessor));
+ }
+
+ return numFastaSeqs;
}
catch(exception& e) {
- m->errorOut(e, "TrimSeqsCommand", "execute");
+ m->errorOut(e, "TrimSeqsCommand", "setLines");
exit(1);
}
}
-
//***************************************************************************************************************
-void TrimSeqsCommand::getOligos(vector<ofstream*>& outFASTAVec){
+void TrimSeqsCommand::getOligos(vector<string>& outFASTAVec){ //vector<ofstream*>& outFASTAVec
try {
ifstream inOligos;
openInputFile(oligoFile, inOligos);
}
else if(type == "barcode"){
inOligos >> group;
- barcodes[oligo]=index++;
+ barcodes[oligo]=index; index++;
groupVector.push_back(group);
if(allFiles){
- outFASTAVec.push_back(new ofstream((outputDir + getRootName(getSimpleName(fastaFile)) + group + ".fasta").c_str(), ios::ate));
- outputNames.push_back((outputDir + getRootName(getSimpleName(fastaFile)) + group + ".fasta"));
+ //outFASTAVec.push_back(new ofstream((outputDir + getRootName(getSimpleName(fastaFile)) + group + ".fasta").c_str(), ios::ate));
+ outputNames.push_back((outputDir + getRootName(getSimpleName(fastaFile)) + toString(index) + "." + group + ".fasta"));
+ outFASTAVec.push_back((outputDir + getRootName(getSimpleName(fastaFile)) + toString(index) + "." + group + ".fasta"));
}
- }
+ }else{ m->mothurOut(type + " is not recognized as a valid type. Choices are forward, reverse, and barcode. Ignoring " + oligo + "."); m->mothurOutEndLine(); }
}
}
m->errorOut(e, "TrimSeqsCommand", "getOligos");
exit(1);
}
-
}
-
//***************************************************************************************************************
-bool TrimSeqsCommand::stripBarcode(Sequence& seq, int& group){
+int TrimSeqsCommand::stripBarcode(Sequence& seq, int& group){
try {
+
string rawSequence = seq.getUnaligned();
- bool success = 0; //guilty until proven innocent
+ 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 = 0;
- break;
+ 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 = 1;
+ success = 0;
break;
}
}
+
+ //if you found the barcode or if you don't want to allow for diffs
+// cout << success;
+ if ((bdiffs == 0) || (success == 0)) { return success; }
+
+ else { //try aligning and see if you can find it
+// cout << endl;
+
+ 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 newStart=0;
+ int numDiff = countDiffs(oligo, temp);
+
+// cout << oligo << '\t' << temp << '\t' << numDiff << endl;
+
+ if(numDiff < minDiff){
+ minDiff = numDiff;
+ minCount = 1;
+ minGroup = it->second;
+ minPos = 0;
+ for(int i=0;i<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; }
+
+ }
+// cout << success << endl;
+
return success;
}
//***************************************************************************************************************
-bool TrimSeqsCommand::stripForward(Sequence& seq){
+int TrimSeqsCommand::stripForward(Sequence& seq){
try {
string rawSequence = seq.getUnaligned();
- bool success = 0; //guilty until proven innocent
+ int success = pdiffs + 1; //guilty until proven innocent
+ //can you find the primer
for(int i=0;i<numFPrimers;i++){
string oligo = forPrimer[i];
-
- if(rawSequence.length() < oligo.length()){
- success = 0;
+
+ if(rawSequence.length() < oligo.length()){
+ success = pdiffs + 1;
break;
}
if(compareDNASeq(oligo, rawSequence.substr(0,oligo.length()))){
seq.setUnaligned(rawSequence.substr(oligo.length()));
- success = 1;
+ success = 0;
break;
}
}
- return success;
+ //if you found the barcode or if you don't want to allow for diffs
+// cout << success;
+ if ((pdiffs == 0) || (success == 0)) { return success; }
+
+ else { //try aligning and see if you can find it
+// cout << endl;
+
+ int maxLength = 0;
+
+ Alignment* alignment;
+ if (numFPrimers > 0) {
+
+ for(int i=0;i<numFPrimers;i++){
+ if(forPrimer[i].length() > maxLength){
+ maxLength = forPrimer[i].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 minPos = 0;
+
+ for(int i=0;i<numFPrimers;i++){
+ string oligo = forPrimer[i];
+
+ 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 newStart=0;
+ int numDiff = countDiffs(oligo, temp);
+ if(numDiff < minDiff){
+ minDiff = numDiff;
+ minCount = 1;
+ minPos = 0;
+ for(int i=0;i<alnLength;i++){
+ if(temp[i] != '-'){
+ minPos++;
+ }
+ }
+ }
+ else if(numDiff == minDiff){
+ minCount++;
+ }
+
+ }
+ if(minDiff > pdiffs) { success = minDiff; }
+ else if(minCount > 1) { success = pdiffs + 10; }
+ else{
+ seq.setUnaligned(rawSequence.substr(minPos));
+ success = minDiff;
+ }
+
+ if (alignment != NULL) { delete alignment; }
+
+ }
+ return success;
+
}
catch(exception& e) {
m->errorOut(e, "TrimSeqsCommand", "stripForward");
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; }
+ 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] == '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; }
+ if(success == 0) { break; }
}
else{
success = 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);
+ }
+
+}
//***************************************************************************************************************
bool TrimSeqsCommand::stripQualThreshold(Sequence& seq, ifstream& qFile){
try {
+// string rawSequence = seq.getUnaligned();
+// int seqLength; // = rawSequence.length();
+// string name, temp, temp2;
+//
+// qFile >> name;
+//
+// //get rest of line
+// temp = "";
+// while (!qFile.eof()) {
+// char c = qFile.get();
+// if (c == 10 || c == 13){ break; }
+// else { temp += c; }
+// }
+//
+// int pos = temp.find("length");
+// if (pos == temp.npos) { m->mothurOut("Cannot find length in qfile for " + seq.getName()); m->mothurOutEndLine(); seqLength = 0; }
+// else {
+// string tempLength = temp.substr(pos);
+// istringstream iss (tempLength,istringstream::in);
+// iss >> temp;
+// }
+//
+// splitAtEquals(temp2, temp); //separates length=242, temp=length, temp2=242
+// convert(temp, seqLength); //converts string to int
+//
+// if (name.length() != 0) { if(name.substr(1) != seq.getName()) { m->mothurOut("sequence name mismatch btwn fasta and qual file"); m->mothurOutEndLine(); } }
+
string rawSequence = seq.getUnaligned();
- int seqLength = rawSequence.length();
+ int seqLength = seq.getNumBases();
+ bool success = 0; //guilty until proven innocent
string name;
qFile >> name;
- if (name.length() != 0) { if(name.substr(1) != seq.getName()) { m->mothurOut("sequence name mismatch btwn fasta and qual file"); m->mothurOutEndLine(); } }
+ if (name[0] == '>') { if(name.substr(1) != seq.getName()) { m->mothurOut("sequence name mismatch btwn fasta: " + seq.getName() + " and qual file: " + name); m->mothurOutEndLine(); } }
+
while (!qFile.eof()) { char c = qFile.get(); if (c == 10 || c == 13){ break; } }
int score;
for(int i=0;i<seqLength;i++){
qFile >> score;
- if(score <= qThreshold){
+ if(score < qThreshold){
end = i;
break;
}
string name;
qFile >> name;
- if (name[0] == '>') { if(name.substr(1) != seq.getName()) { m->mothurOut("sequence name mismatch btwn fasta: " + seq.getName() + " and qual file: " + name); m->mothurOutEndLine(); } }
+ if (name[0] == '>') { if(name.substr(1) != seq.getName()) { m->mothurOut("sequence name mismatch btwn fasta: " + seq.getName() + " and qual file: " + name); m->mothurOutEndLine(); } }
while (!qFile.eof()) { char c = qFile.get(); if (c == 10 || c == 13){ break; } }