5 * Created by Pat Schloss on 6/6/09.
6 * Copyright 2009 Patrick D. Schloss. All rights reserved.
10 #include "trimseqscommand.h"
11 #include "needlemanoverlap.hpp"
13 //***************************************************************************************************************
15 TrimSeqsCommand::TrimSeqsCommand(string option) {
20 //allow user to run help
21 if(option == "help") { help(); abort = true; }
24 //valid paramters for this command
25 string AlignArray[] = {"fasta", "flip", "oligos", "maxambig", "maxhomop", "minlength", "maxlength", "qfile",
26 "qthreshold", "qaverage", "allfiles", "qtrim","tdiffs", "pdiffs", "bdiffs", "processors", "outputdir","inputdir"};
28 vector<string> myArray (AlignArray, AlignArray+(sizeof(AlignArray)/sizeof(string)));
30 OptionParser parser(option);
31 map<string,string> parameters = parser.getParameters();
33 ValidParameters validParameter;
34 map<string,string>::iterator it;
36 //check to make sure all parameters are valid for command
37 for (it = parameters.begin(); it != parameters.end(); it++) {
38 if (validParameter.isValidParameter(it->first, myArray, it->second) != true) { abort = true; }
41 //if the user changes the input directory command factory will send this info to us in the output parameter
42 string inputDir = validParameter.validFile(parameters, "inputdir", false);
43 if (inputDir == "not found"){ inputDir = ""; }
46 it = parameters.find("fasta");
47 //user has given a template file
48 if(it != parameters.end()){
49 path = hasPath(it->second);
50 //if the user has not given a path then, add inputdir. else leave path alone.
51 if (path == "") { parameters["fasta"] = inputDir + it->second; }
54 it = parameters.find("oligos");
55 //user has given a template file
56 if(it != parameters.end()){
57 path = hasPath(it->second);
58 //if the user has not given a path then, add inputdir. else leave path alone.
59 if (path == "") { parameters["oligos"] = inputDir + it->second; }
62 it = parameters.find("qfile");
63 //user has given a template file
64 if(it != parameters.end()){
65 path = hasPath(it->second);
66 //if the user has not given a path then, add inputdir. else leave path alone.
67 if (path == "") { parameters["qfile"] = inputDir + it->second; }
72 //check for required parameters
73 fastaFile = validParameter.validFile(parameters, "fasta", true);
74 if (fastaFile == "not found") { m->mothurOut("fasta is a required parameter for the screen.seqs command."); m->mothurOutEndLine(); abort = true; }
75 else if (fastaFile == "not open") { abort = true; }
77 //if the user changes the output directory command factory will send this info to us in the output parameter
78 outputDir = validParameter.validFile(parameters, "outputdir", false); if (outputDir == "not found"){
80 outputDir += hasPath(fastaFile); //if user entered a file with a path then preserve it
83 //check for optional parameter and set defaults
84 // ...at some point should added some additional type checking...
86 temp = validParameter.validFile(parameters, "flip", false);
87 if (temp == "not found"){ flip = 0; }
88 else if(isTrue(temp)) { flip = 1; }
90 temp = validParameter.validFile(parameters, "oligos", true);
91 if (temp == "not found"){ oligoFile = ""; }
92 else if(temp == "not open"){ abort = true; }
93 else { oligoFile = temp; }
95 temp = validParameter.validFile(parameters, "maxambig", false); if (temp == "not found") { temp = "-1"; }
96 convert(temp, maxAmbig);
98 temp = validParameter.validFile(parameters, "maxhomop", false); if (temp == "not found") { temp = "0"; }
99 convert(temp, maxHomoP);
101 temp = validParameter.validFile(parameters, "minlength", false); if (temp == "not found") { temp = "0"; }
102 convert(temp, minLength);
104 temp = validParameter.validFile(parameters, "maxlength", false); if (temp == "not found") { temp = "0"; }
105 convert(temp, maxLength);
108 temp = validParameter.validFile(parameters, "bdiffs", false); if (temp == "not found") { temp = "0"; }
109 convert(temp, bdiffs);
111 temp = validParameter.validFile(parameters, "pdiffs", false); if (temp == "not found") { temp = "0"; }
112 convert(temp, pdiffs);
114 temp = validParameter.validFile(parameters, "tdiffs", false); if (temp == "not found") { int tempTotal = pdiffs + bdiffs; temp = toString(tempTotal); }
115 convert(temp, tdiffs);
117 if(tdiffs == 0){ tdiffs = bdiffs + pdiffs; }
119 temp = validParameter.validFile(parameters, "qfile", true);
120 if (temp == "not found") { qFileName = ""; }
121 else if(temp == "not open") { abort = true; }
122 else { qFileName = temp; }
124 temp = validParameter.validFile(parameters, "qthreshold", false); if (temp == "not found") { temp = "0"; }
125 convert(temp, qThreshold);
127 temp = validParameter.validFile(parameters, "qtrim", false); if (temp == "not found") { temp = "F"; }
128 qtrim = isTrue(temp);
130 temp = validParameter.validFile(parameters, "qaverage", false); if (temp == "not found") { temp = "0"; }
131 convert(temp, qAverage);
133 temp = validParameter.validFile(parameters, "allfiles", false); if (temp == "not found") { temp = "F"; }
134 allFiles = isTrue(temp);
136 temp = validParameter.validFile(parameters, "processors", false); if (temp == "not found"){ temp = "1"; }
137 convert(temp, processors);
139 if(allFiles && oligoFile == ""){
140 m->mothurOut("You selected allfiles, but didn't enter an oligos file. Ignoring the allfiles request."); m->mothurOutEndLine();
142 if((qAverage != 0 && qThreshold != 0) && qFileName == ""){
143 m->mothurOut("You didn't provide a quality file name, quality criteria will be ignored."); m->mothurOutEndLine();
147 if(!flip && oligoFile=="" && !maxLength && !minLength && (maxAmbig==-1) && !maxHomoP && qFileName == ""){
148 m->mothurOut("You didn't set any options... quiting command."); m->mothurOutEndLine();
154 catch(exception& e) {
155 m->errorOut(e, "TrimSeqsCommand", "TrimSeqsCommand");
159 //**********************************************************************************************************************
161 void TrimSeqsCommand::help(){
163 m->mothurOut("The trim.seqs command reads a fastaFile and creates .....\n");
164 m->mothurOut("The trim.seqs command parameters are fasta, flip, oligos, maxambig, maxhomop, minlength, maxlength, qfile, qthreshold, qaverage, diffs, qtrim and allfiles.\n");
165 m->mothurOut("The fasta parameter is required.\n");
166 m->mothurOut("The flip parameter .... The default is 0.\n");
167 m->mothurOut("The oligos parameter .... The default is "".\n");
168 m->mothurOut("The maxambig parameter .... The default is -1.\n");
169 m->mothurOut("The maxhomop parameter .... The default is 0.\n");
170 m->mothurOut("The minlength parameter .... The default is 0.\n");
171 m->mothurOut("The maxlength parameter .... The default is 0.\n");
172 m->mothurOut("The tdiffs parameter is used to specify the total number of differences allowed in the sequence. The default is pdiffs + bdiffs.\n");
173 m->mothurOut("The bdiffs parameter is used to specify the number of differences allowed in the barcode. The default is 0.\n");
174 m->mothurOut("The pdiffs parameter is used to specify the number of differences allowed in the primer. The default is 0.\n");
175 m->mothurOut("The qfile parameter .....\n");
176 m->mothurOut("The qthreshold parameter .... The default is 0.\n");
177 m->mothurOut("The qaverage parameter .... The default is 0.\n");
178 m->mothurOut("The allfiles parameter .... The default is F.\n");
179 m->mothurOut("The qtrim parameter .... The default is F.\n");
180 m->mothurOut("The trim.seqs command should be in the following format: \n");
181 m->mothurOut("trim.seqs(fasta=yourFastaFile, flip=yourFlip, oligos=yourOligos, maxambig=yourMaxambig, \n");
182 m->mothurOut("maxhomop=yourMaxhomop, minlength=youMinlength, maxlength=yourMaxlength) \n");
183 m->mothurOut("Example trim.seqs(fasta=abrecovery.fasta, flip=..., oligos=..., maxambig=..., maxhomop=..., minlength=..., maxlength=...).\n");
184 m->mothurOut("Note: No spaces between parameter labels (i.e. fasta), '=' and parameters (i.e.yourFasta).\n");
185 m->mothurOut("For more details please check out the wiki http://www.mothur.org/wiki/Trim.seqs .\n\n");
188 catch(exception& e) {
189 m->errorOut(e, "TrimSeqsCommand", "help");
195 //***************************************************************************************************************
197 TrimSeqsCommand::~TrimSeqsCommand(){ /* do nothing */ }
199 //***************************************************************************************************************
201 int TrimSeqsCommand::execute(){
204 if (abort == true) { return 0; }
206 numFPrimers = 0; //this needs to be initialized
209 string trimSeqFile = outputDir + getRootName(getSimpleName(fastaFile)) + "trim.fasta";
210 outputNames.push_back(trimSeqFile);
211 string scrapSeqFile = outputDir + getRootName(getSimpleName(fastaFile)) + "scrap.fasta";
212 outputNames.push_back(scrapSeqFile);
213 string groupFile = outputDir + getRootName(getSimpleName(fastaFile)) + "groups";
215 vector<string> fastaFileNames;
217 outputNames.push_back(groupFile);
218 getOligos(fastaFileNames);
221 if(qFileName != "") { setLines(qFileName, qLines); }
224 #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
227 openInputFile(fastaFile, inFASTA);
228 int numSeqs=count(istreambuf_iterator<char>(inFASTA),istreambuf_iterator<char>(), '>');
231 lines.push_back(new linePair(0, numSeqs));
233 driverCreateTrim(fastaFile, qFileName, trimSeqFile, scrapSeqFile, groupFile, fastaFileNames, lines[0], lines[0]);
235 for (int j = 0; j < fastaFileNames.size(); j++) {
236 rename((fastaFileNames[j] + toString(getpid()) + ".temp").c_str(), fastaFileNames[j].c_str());
240 setLines(fastaFile, lines);
241 if(qFileName == "") { qLines = lines; }
243 createProcessesCreateTrim(fastaFile, qFileName, trimSeqFile, scrapSeqFile, groupFile, fastaFileNames);
245 rename((trimSeqFile + toString(processIDS[0]) + ".temp").c_str(), trimSeqFile.c_str());
246 rename((scrapSeqFile + toString(processIDS[0]) + ".temp").c_str(), scrapSeqFile.c_str());
247 rename((groupFile + toString(processIDS[0]) + ".temp").c_str(), groupFile.c_str());
248 for (int j = 0; j < fastaFileNames.size(); j++) {
249 rename((fastaFileNames[j] + toString(processIDS[0]) + ".temp").c_str(), fastaFileNames[j].c_str());
252 for(int i=1;i<processors;i++){
253 appendFiles((trimSeqFile + toString(processIDS[i]) + ".temp"), trimSeqFile);
254 remove((trimSeqFile + toString(processIDS[i]) + ".temp").c_str());
255 appendFiles((scrapSeqFile + toString(processIDS[i]) + ".temp"), scrapSeqFile);
256 remove((scrapSeqFile + toString(processIDS[i]) + ".temp").c_str());
257 appendFiles((groupFile + toString(processIDS[i]) + ".temp"), groupFile);
258 remove((groupFile + toString(processIDS[i]) + ".temp").c_str());
259 for (int j = 0; j < fastaFileNames.size(); j++) {
260 appendFiles((fastaFileNames[j] + toString(processIDS[i]) + ".temp"), fastaFileNames[j]);
261 remove((fastaFileNames[j] + toString(processIDS[i]) + ".temp").c_str());
266 if (m->control_pressed) { return 0; }
269 openInputFile(fastafileNames[s], inFASTA);
270 numSeqs=count(istreambuf_iterator<char>(inFASTA),istreambuf_iterator<char>(), '>');
273 lines.push_back(new linePair(0, numSeqs));
275 driverCreateSummary(fastafile, qFileName, trimSeqFile, scrapSeqFile, groupFile, fastaFileNames, lines[0], lines[0]);
277 if (m->control_pressed) { return 0; }
281 for(int i=0;i<fastaFileNames.size();i++){
284 openInputFile(getRootName(fastaFile) + groupVector[i] + ".fasta", inFASTA);
286 openOutputFile(outputDir + getRootName(getSimpleName(fastaFile)) + groupVector[i] + ".groups", outGroups);
287 outputNames.push_back(outputDir + getRootName(getSimpleName(fastaFile)) + groupVector[i] + ".groups");
289 while(!inFASTA.eof()){
290 if(inFASTA.get() == '>'){
292 outGroups << seqName << '\t' << groupVector[i] << endl;
294 while (!inFASTA.eof()) { char c = inFASTA.get(); if (c == 10 || c == 13){ break; } }
300 if (m->control_pressed) {
301 for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); }
305 m->mothurOutEndLine();
306 m->mothurOut("Output File Names: "); m->mothurOutEndLine();
307 for (int i = 0; i < outputNames.size(); i++) { m->mothurOut(outputNames[i]); m->mothurOutEndLine(); }
308 m->mothurOutEndLine();
313 catch(exception& e) {
314 m->errorOut(e, "TrimSeqsCommand", "execute");
319 /**************************************************************************************/
320 int TrimSeqsCommand::driverCreateTrim(string filename, string qFileName, string trimFile, string scrapFile, string groupFile, vector<string> fastaNames, linePair* line, linePair* qline) {
324 int able = openOutputFile(trimFile, outFASTA);
327 openOutputFile(scrapFile, scrapFASTA);
330 vector<ofstream*> fastaFileNames;
331 if (oligoFile != "") {
332 openOutputFile(groupFile, outGroups);
333 for (int i = 0; i < fastaNames.size(); i++) {
334 fastaFileNames.push_back(new ofstream((fastaNames[i] + toString(getpid()) + ".temp").c_str(), ios::ate));
339 openInputFile(filename, inFASTA);
342 if(qFileName != "") { openInputFile(qFileName, qFile); }
344 qFile.seekg(qline->start);
345 inFASTA.seekg(line->start);
347 for(int i=0;i<line->num;i++){
349 if (m->control_pressed) {
353 if (oligoFile != "") { outGroups.close(); }
354 if(qFileName != "") { qFile.close(); }
355 for(int i=0;i<fastaFileNames.size();i++){
356 fastaFileNames[i]->close();
357 delete fastaFileNames[i];
359 for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); }
365 Sequence currSeq(inFASTA);
367 string origSeq = currSeq.getUnaligned();
370 string trashCode = "";
371 int currentSeqsDiffs = 0;
374 if(qThreshold != 0) { success = stripQualThreshold(currSeq, qFile); }
375 else if(qAverage != 0) { success = cullQualAverage(currSeq, qFile); }
376 if ((!qtrim) && (origSeq.length() != currSeq.getUnaligned().length())) {
377 success = 0; //if you don't want to trim and the sequence does not meet quality requirements, move to scrap
379 if(!success) { trashCode += 'q'; }
382 if(barcodes.size() != 0){
383 success = stripBarcode(currSeq, group);
384 // cout << "here: " << success << endl;
385 if(success > bdiffs){ trashCode += 'b'; }
386 else{ currentSeqsDiffs += success; }
389 if(numFPrimers != 0){
390 success = stripForward(currSeq);
391 if(success > pdiffs){ trashCode += 'f'; }
392 else{ currentSeqsDiffs += success; }
395 if (currentSeqsDiffs > tdiffs) { trashCode += 't'; }
397 if(numRPrimers != 0){
398 success = stripReverse(currSeq);
399 if(!success){ trashCode += 'r'; }
402 if(minLength > 0 || maxLength > 0){
403 success = cullLength(currSeq);
404 if(!success){ trashCode += 'l'; }
407 success = cullHomoP(currSeq);
408 if(!success){ trashCode += 'h'; }
411 success = cullAmbigs(currSeq);
412 if(!success){ trashCode += 'n'; }
415 if(flip){ currSeq.reverseComplement(); } // should go last
417 if(trashCode.length() == 0){
418 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.
419 currSeq.printSequence(outFASTA);
420 if(barcodes.size() != 0){
421 outGroups << currSeq.getName() << '\t' << groupVector[group] << endl;
424 currSeq.printSequence(*fastaFileNames[group]);
429 currSeq.setName(currSeq.getName() + '|' + trashCode);
430 currSeq.setUnaligned(origSeq);
431 currSeq.printSequence(scrapFASTA);
440 if (oligoFile != "") { outGroups.close(); }
441 if(qFileName != "") { qFile.close(); }
443 for(int i=0;i<fastaFileNames.size();i++){
444 fastaFileNames[i]->close();
445 delete fastaFileNames[i];
450 catch(exception& e) {
451 m->errorOut(e, "TrimSeqsCommand", "driverCreateTrim");
455 /**************************************************************************************************/
456 int TrimSeqsCommand::createProcessesCreateTrim(string filename, string qFileName, string trimFile, string scrapFile, string groupFile, vector<string> fastaNames) {
458 #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
463 //loop through and create all the processes you want
464 while (process != processors) {
468 processIDS.push_back(pid); //create map from line number to pid so you can append files in correct order later
471 driverCreateTrim(filename, qFileName, (trimFile + toString(getpid()) + ".temp"), (scrapFile + toString(getpid()) + ".temp"), (groupFile + toString(getpid()) + ".temp"), fastaNames, lines[process], qLines[process]);
473 }else { m->mothurOut("unable to spawn the necessary processes."); m->mothurOutEndLine(); exit(0); }
476 //force parent to wait until all the processes are done
477 for (int i=0;i<processors;i++) {
478 int temp = processIDS[i];
485 catch(exception& e) {
486 m->errorOut(e, "TrimSeqsCommand", "createProcessesCreateTrim");
490 /**************************************************************************************************/
492 int TrimSeqsCommand::setLines(string filename, vector<linePair*>& lines) {
497 vector<long int> positions;
500 openInputFile(filename, inFASTA);
503 while(!inFASTA.eof()){
504 input = getline(inFASTA);
506 if (input.length() != 0) {
507 if(input[0] == '>'){ long int pos = inFASTA.tellg(); positions.push_back(pos - input.length() - 1); }
512 int numFastaSeqs = positions.size();
517 //get num bytes in file
518 pFile = fopen (filename.c_str(),"rb");
519 if (pFile==NULL) perror ("Error opening file");
521 fseek (pFile, 0, SEEK_END);
526 int numSeqsPerProcessor = numFastaSeqs / processors;
528 for (int i = 0; i < processors; i++) {
530 long int startPos = positions[ i * numSeqsPerProcessor ];
531 if(i == processors - 1){
532 numSeqsPerProcessor = numFastaSeqs - i * numSeqsPerProcessor;
534 long int myEnd = positions[ (i+1) * numSeqsPerProcessor ];
536 lines.push_back(new linePair(startPos, numSeqsPerProcessor));
541 catch(exception& e) {
542 m->errorOut(e, "TrimSeqsCommand", "setLines");
546 //***************************************************************************************************************
548 void TrimSeqsCommand::getOligos(vector<string>& outFASTAVec){ //vector<ofstream*>& outFASTAVec
551 openInputFile(oligoFile, inOligos);
555 string type, oligo, group;
558 while(!inOligos.eof()){
562 while (!inOligos.eof()) { char c = inOligos.get(); if (c == 10 || c == 13){ break; } } // get rest of line if there's any crap there
567 for(int i=0;i<oligo.length();i++){
568 oligo[i] = toupper(oligo[i]);
569 if(oligo[i] == 'U') { oligo[i] = 'T'; }
572 if(type == "forward"){
573 forPrimer.push_back(oligo);
575 else if(type == "reverse"){
576 Sequence oligoRC("reverse", oligo);
577 oligoRC.reverseComplement();
578 revPrimer.push_back(oligoRC.getUnaligned());
580 else if(type == "barcode"){
582 barcodes[oligo]=index++;
583 groupVector.push_back(group);
586 //outFASTAVec.push_back(new ofstream((outputDir + getRootName(getSimpleName(fastaFile)) + group + ".fasta").c_str(), ios::ate));
587 outputNames.push_back((outputDir + getRootName(getSimpleName(fastaFile)) + group + ".fasta"));
588 outFASTAVec.push_back((outputDir + getRootName(getSimpleName(fastaFile)) + group + ".fasta"));
596 numFPrimers = forPrimer.size();
597 numRPrimers = revPrimer.size();
600 catch(exception& e) {
601 m->errorOut(e, "TrimSeqsCommand", "getOligos");
605 //***************************************************************************************************************
607 bool TrimSeqsCommand::stripBarcode(Sequence& seq, int& group){
609 string rawSequence = seq.getUnaligned();
610 bool success = bdiffs + 1; //guilty until proven innocent
612 //can you find the barcode
613 for(map<string,int>::iterator it=barcodes.begin();it!=barcodes.end();it++){
614 string oligo = it->first;
615 if(rawSequence.length() < oligo.length()){ //let's just assume that the barcodes are the same length
616 success = bdiffs + 1;
620 if(compareDNASeq(oligo, rawSequence.substr(0,oligo.length()))){
622 seq.setUnaligned(rawSequence.substr(oligo.length()));
628 //if you found the barcode or if you don't want to allow for diffs
630 if ((bdiffs == 0) || (success == 0)) { return success; }
632 else { //try aligning and see if you can find it
637 Alignment* alignment;
638 if (barcodes.size() > 0) {
639 map<string,int>::iterator it=barcodes.begin();
641 for(it;it!=barcodes.end();it++){
642 if(it->first.length() > maxLength){
643 maxLength = it->first.length();
646 alignment = new NeedlemanOverlap(-2.0, 1.0, -1.0, (maxLength+bdiffs+1));
648 }else{ alignment = NULL; }
650 //can you find the barcode
656 for(map<string,int>::iterator it=barcodes.begin();it!=barcodes.end();it++){
657 string oligo = it->first;
658 // int length = oligo.length();
660 if(rawSequence.length() < maxLength){ //let's just assume that the barcodes are the same length
661 success = bdiffs + 1;
665 //use needleman to align first barcode.length()+numdiffs of sequence to each barcode
666 alignment->align(oligo, rawSequence.substr(0,oligo.length()+bdiffs));
667 oligo = alignment->getSeqAAln();
668 string temp = alignment->getSeqBAln();
670 int alnLength = oligo.length();
672 for(int i=oligo.length()-1;i>=0;i--){
673 if(oligo[i] != '-'){ alnLength = i+1; break; }
675 oligo = oligo.substr(0,alnLength);
676 temp = temp.substr(0,alnLength);
677 // cout << "barcode = " << oligo << " raw = " << rawSequence.substr(0,alnLength) << " raw aligned = " << temp << endl;
678 cout << seq.getName() << endl;
679 cout << temp << endl;
680 cout << oligo << endl;
681 cout << alnLength << endl;
685 int numDiff = countDiffs(oligo, temp);
686 if(numDiff < minDiff){
689 minGroup = it->second;
691 for(int i=0;i<alnLength;i++){
697 else if(numDiff == minDiff){
702 if(minDiff > bdiffs){ success = bdiffs + 1; }
703 else if(minCount > 1) { success = bdiffs + 1; }
706 seq.setUnaligned("*" + rawSequence.substr(minPos));
710 if (alignment != NULL) { delete alignment; }
716 catch(exception& e) {
717 m->errorOut(e, "TrimSeqsCommand", "stripBarcode");
723 //***************************************************************************************************************
725 bool TrimSeqsCommand::stripForward(Sequence& seq){
727 string rawSequence = seq.getUnaligned();
728 bool success = pdiffs + 1; //guilty until proven innocent
730 //can you find the primer
731 for(int i=0;i<numFPrimers;i++){
732 string oligo = forPrimer[i];
734 if(rawSequence.length() < oligo.length()){
735 success = pdiffs + 1;
739 if(compareDNASeq(oligo, rawSequence.substr(0,oligo.length()))){
740 seq.setUnaligned(rawSequence.substr(oligo.length()));
746 //if you found the barcode or if you don't want to allow for diffs
748 if ((pdiffs == 0) || (success == 0)) { return success; }
750 else { //try aligning and see if you can find it
755 Alignment* alignment;
756 if (numFPrimers > 0) {
758 for(int i=0;i<numFPrimers;i++){
759 if(forPrimer[i].length() > maxLength){
760 maxLength = forPrimer[i].length();
763 alignment = new NeedlemanOverlap(-2.0, 1.0, -1.0, (maxLength+pdiffs+1));
765 }else{ alignment = NULL; }
767 //can you find the barcode
772 for(int i=0;i<numFPrimers;i++){
773 string oligo = forPrimer[i];
775 if(rawSequence.length() < maxLength){
776 success = pdiffs + 1;
780 //use needleman to align first barcode.length()+numdiffs of sequence to each barcode
781 alignment->align(oligo, rawSequence.substr(0,oligo.length()+pdiffs));
782 oligo = alignment->getSeqAAln();
783 string temp = alignment->getSeqBAln();
785 int alnLength = oligo.length();
787 for(int i=oligo.length()-1;i>=0;i--){
788 if(oligo[i] != '-'){ alnLength = i+1; break; }
790 oligo = oligo.substr(0,alnLength);
791 temp = temp.substr(0,alnLength);
792 // cout << "barcode = " << oligo << " raw = " << rawSequence.substr(0,alnLength) << " raw aligned = " << temp << endl;
793 cout << seq.getName() << endl;
794 cout << temp << endl;
795 cout << oligo << endl;
796 cout << alnLength << endl;
800 int numDiff = countDiffs(oligo, temp);
801 if(numDiff < minDiff){
805 for(int i=0;i<alnLength;i++){
811 else if(numDiff == minDiff){
816 if(minDiff > pdiffs){ success = pdiffs + 1; }
817 else if(minCount > 1) { success = pdiffs + 1; }
819 seq.setUnaligned("*" + rawSequence.substr(minPos));
823 if (alignment != NULL) { delete alignment; }
829 catch(exception& e) {
830 m->errorOut(e, "TrimSeqsCommand", "stripForward");
835 //***************************************************************************************************************
837 bool TrimSeqsCommand::stripReverse(Sequence& seq){
839 string rawSequence = seq.getUnaligned();
840 bool success = 0; //guilty until proven innocent
842 for(int i=0;i<numRPrimers;i++){
843 string oligo = revPrimer[i];
845 if(rawSequence.length() < oligo.length()){
850 if(compareDNASeq(oligo, rawSequence.substr(rawSequence.length()-oligo.length(),oligo.length()))){
851 seq.setUnaligned(rawSequence.substr(0,rawSequence.length()-oligo.length()));
859 catch(exception& e) {
860 m->errorOut(e, "TrimSeqsCommand", "stripReverse");
865 //***************************************************************************************************************
867 bool TrimSeqsCommand::cullLength(Sequence& seq){
870 int length = seq.getNumBases();
871 bool success = 0; //guilty until proven innocent
873 if(length >= minLength && maxLength == 0) { success = 1; }
874 else if(length >= minLength && length <= maxLength) { success = 1; }
875 else { success = 0; }
880 catch(exception& e) {
881 m->errorOut(e, "TrimSeqsCommand", "cullLength");
887 //***************************************************************************************************************
889 bool TrimSeqsCommand::cullHomoP(Sequence& seq){
891 int longHomoP = seq.getLongHomoPolymer();
892 bool success = 0; //guilty until proven innocent
894 if(longHomoP <= maxHomoP){ success = 1; }
895 else { success = 0; }
899 catch(exception& e) {
900 m->errorOut(e, "TrimSeqsCommand", "cullHomoP");
906 //***************************************************************************************************************
908 bool TrimSeqsCommand::cullAmbigs(Sequence& seq){
910 int numNs = seq.getAmbigBases();
911 bool success = 0; //guilty until proven innocent
913 if(numNs <= maxAmbig) { success = 1; }
914 else { success = 0; }
918 catch(exception& e) {
919 m->errorOut(e, "TrimSeqsCommand", "cullAmbigs");
925 //***************************************************************************************************************
927 bool TrimSeqsCommand::compareDNASeq(string oligo, string seq){
930 int length = oligo.length();
932 for(int i=0;i<length;i++){
934 if(oligo[i] != seq[i]){
935 if(oligo[i] == 'A' || oligo[i] == 'T' || oligo[i] == 'G' || oligo[i] == 'C') { success = 0; }
936 else if((oligo[i] == 'N' || oligo[i] == 'I') && (seq[i] == 'N')) { success = 0; }
937 else if(oligo[i] == 'R' && (seq[i] != 'A' && seq[i] != 'G')) { success = 0; }
938 else if(oligo[i] == 'Y' && (seq[i] != 'C' && seq[i] != 'T')) { success = 0; }
939 else if(oligo[i] == 'M' && (seq[i] != 'C' && seq[i] != 'A')) { success = 0; }
940 else if(oligo[i] == 'K' && (seq[i] != 'T' && seq[i] != 'G')) { success = 0; }
941 else if(oligo[i] == 'W' && (seq[i] != 'T' && seq[i] != 'A')) { success = 0; }
942 else if(oligo[i] == 'S' && (seq[i] != 'C' && seq[i] != 'G')) { success = 0; }
943 else if(oligo[i] == 'B' && (seq[i] != 'C' && seq[i] != 'T' && seq[i] != 'G')) { success = 0; }
944 else if(oligo[i] == 'D' && (seq[i] != 'A' && seq[i] != 'T' && seq[i] != 'G')) { success = 0; }
945 else if(oligo[i] == 'H' && (seq[i] != 'A' && seq[i] != 'T' && seq[i] != 'C')) { success = 0; }
946 else if(oligo[i] == 'V' && (seq[i] != 'A' && seq[i] != 'C' && seq[i] != 'G')) { success = 0; }
948 if(success == 0) { break; }
957 catch(exception& e) {
958 m->errorOut(e, "TrimSeqsCommand", "compareDNASeq");
963 //***************************************************************************************************************
965 int TrimSeqsCommand::countDiffs(string oligo, string seq){
968 int length = oligo.length();
971 for(int i=0;i<length;i++){
973 if(oligo[i] != seq[i]){
974 if(oligo[i] == 'A' || oligo[i] == 'T' || oligo[i] == 'G' || oligo[i] == 'C' || oligo[i] == '-' || oligo[i] == '.') { countDiffs++; }
975 else if((oligo[i] == 'N' || oligo[i] == 'I') && (seq[i] == 'N')) { countDiffs++; }
976 else if(oligo[i] == 'R' && (seq[i] != 'A' && seq[i] != 'G')) { countDiffs++; }
977 else if(oligo[i] == 'Y' && (seq[i] != 'C' && seq[i] != 'T')) { countDiffs++; }
978 else if(oligo[i] == 'M' && (seq[i] != 'C' && seq[i] != 'A')) { countDiffs++; }
979 else if(oligo[i] == 'K' && (seq[i] != 'T' && seq[i] != 'G')) { countDiffs++; }
980 else if(oligo[i] == 'W' && (seq[i] != 'T' && seq[i] != 'A')) { countDiffs++; }
981 else if(oligo[i] == 'S' && (seq[i] != 'C' && seq[i] != 'G')) { countDiffs++; }
982 else if(oligo[i] == 'B' && (seq[i] != 'C' && seq[i] != 'T' && seq[i] != 'G')) { countDiffs++; }
983 else if(oligo[i] == 'D' && (seq[i] != 'A' && seq[i] != 'T' && seq[i] != 'G')) { countDiffs++; }
984 else if(oligo[i] == 'H' && (seq[i] != 'A' && seq[i] != 'T' && seq[i] != 'C')) { countDiffs++; }
985 else if(oligo[i] == 'V' && (seq[i] != 'A' && seq[i] != 'C' && seq[i] != 'G')) { countDiffs++; }
992 catch(exception& e) {
993 m->errorOut(e, "TrimSeqsCommand", "countDiffs");
998 //***************************************************************************************************************
1000 bool TrimSeqsCommand::stripQualThreshold(Sequence& seq, ifstream& qFile){
1002 string rawSequence = seq.getUnaligned();
1003 int seqLength; // = rawSequence.length();
1004 string name, temp, temp2;
1006 qFile >> name >> temp;
1008 splitAtEquals(temp2, temp); //separates length=242, temp=length, temp2=242
1009 convert(temp, seqLength); //converts string to int
1011 if (name.length() != 0) { if(name.substr(1) != seq.getName()) { m->mothurOut("sequence name mismatch btwn fasta and qual file"); m->mothurOutEndLine(); } }
1012 while (!qFile.eof()) { char c = qFile.get(); if (c == 10 || c == 13){ break; } }
1015 int end = seqLength;
1017 for(int i=0;i<seqLength;i++){
1020 if(score <= qThreshold){
1025 for(int i=end+1;i<seqLength;i++){
1029 seq.setUnaligned(rawSequence.substr(0,end));
1033 catch(exception& e) {
1034 m->errorOut(e, "TrimSeqsCommand", "stripQualThreshold");
1039 //***************************************************************************************************************
1041 bool TrimSeqsCommand::cullQualAverage(Sequence& seq, ifstream& qFile){
1043 string rawSequence = seq.getUnaligned();
1044 int seqLength = seq.getNumBases();
1045 bool success = 0; //guilty until proven innocent
1049 if (name[0] == '>') { if(name.substr(1) != seq.getName()) { m->mothurOut("sequence name mismatch btwn fasta: " + seq.getName() + " and qual file: " + name); m->mothurOutEndLine(); } }
1051 while (!qFile.eof()) { char c = qFile.get(); if (c == 10 || c == 13){ break; } }
1056 for(int i=0;i<seqLength;i++){
1060 average /= seqLength;
1062 if(average >= qAverage) { success = 1; }
1063 else { success = 0; }
1067 catch(exception& e) {
1068 m->errorOut(e, "TrimSeqsCommand", "cullQualAverage");
1073 //***************************************************************************************************************