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) {
21 //allow user to run help
22 if(option == "help") { help(); abort = true; }
25 //valid paramters for this command
26 string AlignArray[] = {"fasta", "flip", "oligos", "maxambig", "maxhomop", "minlength", "maxlength", "qfile",
27 "qthreshold", "qaverage", "allfiles", "qtrim","tdiffs", "pdiffs", "bdiffs", "processors", "outputdir","inputdir"};
29 vector<string> myArray (AlignArray, AlignArray+(sizeof(AlignArray)/sizeof(string)));
31 OptionParser parser(option);
32 map<string,string> parameters = parser.getParameters();
34 ValidParameters validParameter;
35 map<string,string>::iterator it;
37 //check to make sure all parameters are valid for command
38 for (it = parameters.begin(); it != parameters.end(); it++) {
39 if (validParameter.isValidParameter(it->first, myArray, it->second) != true) { abort = true; }
42 //if the user changes the input directory command factory will send this info to us in the output parameter
43 string inputDir = validParameter.validFile(parameters, "inputdir", false);
44 if (inputDir == "not found"){ inputDir = ""; }
47 it = parameters.find("fasta");
48 //user has given a template file
49 if(it != parameters.end()){
50 path = hasPath(it->second);
51 //if the user has not given a path then, add inputdir. else leave path alone.
52 if (path == "") { parameters["fasta"] = inputDir + it->second; }
55 it = parameters.find("oligos");
56 //user has given a template file
57 if(it != parameters.end()){
58 path = hasPath(it->second);
59 //if the user has not given a path then, add inputdir. else leave path alone.
60 if (path == "") { parameters["oligos"] = inputDir + it->second; }
63 it = parameters.find("qfile");
64 //user has given a template file
65 if(it != parameters.end()){
66 path = hasPath(it->second);
67 //if the user has not given a path then, add inputdir. else leave path alone.
68 if (path == "") { parameters["qfile"] = inputDir + it->second; }
73 //check for required parameters
74 fastaFile = validParameter.validFile(parameters, "fasta", true);
75 if (fastaFile == "not found") { m->mothurOut("fasta is a required parameter for the screen.seqs command."); m->mothurOutEndLine(); abort = true; }
76 else if (fastaFile == "not open") { abort = true; }
78 //if the user changes the output directory command factory will send this info to us in the output parameter
79 outputDir = validParameter.validFile(parameters, "outputdir", false); if (outputDir == "not found"){
81 outputDir += hasPath(fastaFile); //if user entered a file with a path then preserve it
84 //check for optional parameter and set defaults
85 // ...at some point should added some additional type checking...
87 temp = validParameter.validFile(parameters, "flip", false);
88 if (temp == "not found"){ flip = 0; }
89 else if(isTrue(temp)) { flip = 1; }
91 temp = validParameter.validFile(parameters, "oligos", true);
92 if (temp == "not found"){ oligoFile = ""; }
93 else if(temp == "not open"){ abort = true; }
94 else { oligoFile = temp; }
96 temp = validParameter.validFile(parameters, "maxambig", false); if (temp == "not found") { temp = "-1"; }
97 convert(temp, maxAmbig);
99 temp = validParameter.validFile(parameters, "maxhomop", false); if (temp == "not found") { temp = "0"; }
100 convert(temp, maxHomoP);
102 temp = validParameter.validFile(parameters, "minlength", false); if (temp == "not found") { temp = "0"; }
103 convert(temp, minLength);
105 temp = validParameter.validFile(parameters, "maxlength", false); if (temp == "not found") { temp = "0"; }
106 convert(temp, maxLength);
109 temp = validParameter.validFile(parameters, "bdiffs", false); if (temp == "not found") { temp = "0"; }
110 convert(temp, bdiffs);
112 temp = validParameter.validFile(parameters, "pdiffs", false); if (temp == "not found") { temp = "0"; }
113 convert(temp, pdiffs);
115 temp = validParameter.validFile(parameters, "tdiffs", false); if (temp == "not found") { int tempTotal = pdiffs + bdiffs; temp = toString(tempTotal); }
116 convert(temp, tdiffs);
118 if(tdiffs == 0){ tdiffs = bdiffs + pdiffs; }
120 temp = validParameter.validFile(parameters, "qfile", true);
121 if (temp == "not found") { qFileName = ""; }
122 else if(temp == "not open") { abort = true; }
123 else { qFileName = temp; }
125 temp = validParameter.validFile(parameters, "qthreshold", false); if (temp == "not found") { temp = "0"; }
126 convert(temp, qThreshold);
128 temp = validParameter.validFile(parameters, "qtrim", false); if (temp == "not found") { temp = "F"; }
129 qtrim = isTrue(temp);
131 temp = validParameter.validFile(parameters, "qaverage", false); if (temp == "not found") { temp = "0"; }
132 convert(temp, qAverage);
134 temp = validParameter.validFile(parameters, "allfiles", false); if (temp == "not found") { temp = "F"; }
135 allFiles = isTrue(temp);
137 temp = validParameter.validFile(parameters, "processors", false); if (temp == "not found"){ temp = "1"; }
138 convert(temp, processors);
140 if(allFiles && oligoFile == ""){
141 m->mothurOut("You selected allfiles, but didn't enter an oligos file. Ignoring the allfiles request."); m->mothurOutEndLine();
143 if((qAverage != 0 && qThreshold != 0) && qFileName == ""){
144 m->mothurOut("You didn't provide a quality file name, quality criteria will be ignored."); m->mothurOutEndLine();
148 if(!flip && oligoFile=="" && !maxLength && !minLength && (maxAmbig==-1) && !maxHomoP && qFileName == ""){
149 m->mothurOut("You didn't set any options... quiting command."); m->mothurOutEndLine();
155 catch(exception& e) {
156 m->errorOut(e, "TrimSeqsCommand", "TrimSeqsCommand");
160 //**********************************************************************************************************************
162 void TrimSeqsCommand::help(){
164 m->mothurOut("The trim.seqs command reads a fastaFile and creates .....\n");
165 m->mothurOut("The trim.seqs command parameters are fasta, flip, oligos, maxambig, maxhomop, minlength, maxlength, qfile, qthreshold, qaverage, diffs, qtrim and allfiles.\n");
166 m->mothurOut("The fasta parameter is required.\n");
167 m->mothurOut("The flip parameter will output the reverse compliment of your trimmed sequence. The default is false.\n");
168 m->mothurOut("The oligos parameter .... The default is "".\n");
169 m->mothurOut("The maxambig parameter .... The default is -1.\n");
170 m->mothurOut("The maxhomop parameter .... The default is 0.\n");
171 m->mothurOut("The minlength parameter .... The default is 0.\n");
172 m->mothurOut("The maxlength parameter .... The default is 0.\n");
173 m->mothurOut("The tdiffs parameter is used to specify the total number of differences allowed in the sequence. The default is pdiffs + bdiffs.\n");
174 m->mothurOut("The bdiffs parameter is used to specify the number of differences allowed in the barcode. The default is 0.\n");
175 m->mothurOut("The pdiffs parameter is used to specify the number of differences allowed in the primer. The default is 0.\n");
176 m->mothurOut("The qfile parameter .....\n");
177 m->mothurOut("The qthreshold parameter .... The default is 0.\n");
178 m->mothurOut("The qaverage parameter .... The default is 0.\n");
179 m->mothurOut("The allfiles parameter .... The default is F.\n");
180 m->mothurOut("The qtrim parameter .... The default is F.\n");
181 m->mothurOut("The trim.seqs command should be in the following format: \n");
182 m->mothurOut("trim.seqs(fasta=yourFastaFile, flip=yourFlip, oligos=yourOligos, maxambig=yourMaxambig, \n");
183 m->mothurOut("maxhomop=yourMaxhomop, minlength=youMinlength, maxlength=yourMaxlength) \n");
184 m->mothurOut("Example trim.seqs(fasta=abrecovery.fasta, flip=..., oligos=..., maxambig=..., maxhomop=..., minlength=..., maxlength=...).\n");
185 m->mothurOut("Note: No spaces between parameter labels (i.e. fasta), '=' and parameters (i.e.yourFasta).\n");
186 m->mothurOut("For more details please check out the wiki http://www.mothur.org/wiki/Trim.seqs .\n\n");
189 catch(exception& e) {
190 m->errorOut(e, "TrimSeqsCommand", "help");
196 //***************************************************************************************************************
198 TrimSeqsCommand::~TrimSeqsCommand(){ /* do nothing */ }
200 //***************************************************************************************************************
202 int TrimSeqsCommand::execute(){
205 if (abort == true) { return 0; }
207 numFPrimers = 0; //this needs to be initialized
210 string trimSeqFile = outputDir + getRootName(getSimpleName(fastaFile)) + "trim.fasta";
211 outputNames.push_back(trimSeqFile);
212 string scrapSeqFile = outputDir + getRootName(getSimpleName(fastaFile)) + "scrap.fasta";
213 outputNames.push_back(scrapSeqFile);
214 string groupFile = outputDir + getRootName(getSimpleName(fastaFile)) + "groups";
216 vector<string> fastaFileNames;
218 outputNames.push_back(groupFile);
219 getOligos(fastaFileNames);
222 if(qFileName != "") { setLines(qFileName, qLines); }
225 #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
229 openInputFile(fastaFile, inFASTA);
230 getNumSeqs(inFASTA, numSeqs);
233 lines.push_back(new linePair(0, numSeqs));
235 driverCreateTrim(fastaFile, qFileName, trimSeqFile, scrapSeqFile, groupFile, fastaFileNames, lines[0], lines[0]);
237 for (int j = 0; j < fastaFileNames.size(); j++) {
238 rename((fastaFileNames[j] + toString(getpid()) + ".temp").c_str(), fastaFileNames[j].c_str());
242 setLines(fastaFile, lines);
243 if(qFileName == "") { qLines = lines; }
245 createProcessesCreateTrim(fastaFile, qFileName, trimSeqFile, scrapSeqFile, groupFile, fastaFileNames);
247 rename((trimSeqFile + toString(processIDS[0]) + ".temp").c_str(), trimSeqFile.c_str());
248 rename((scrapSeqFile + toString(processIDS[0]) + ".temp").c_str(), scrapSeqFile.c_str());
249 rename((groupFile + toString(processIDS[0]) + ".temp").c_str(), groupFile.c_str());
250 for (int j = 0; j < fastaFileNames.size(); j++) {
251 rename((fastaFileNames[j] + toString(processIDS[0]) + ".temp").c_str(), fastaFileNames[j].c_str());
254 for(int i=1;i<processors;i++){
255 appendFiles((trimSeqFile + toString(processIDS[i]) + ".temp"), trimSeqFile);
256 remove((trimSeqFile + toString(processIDS[i]) + ".temp").c_str());
257 appendFiles((scrapSeqFile + toString(processIDS[i]) + ".temp"), scrapSeqFile);
258 remove((scrapSeqFile + toString(processIDS[i]) + ".temp").c_str());
259 appendFiles((groupFile + toString(processIDS[i]) + ".temp"), groupFile);
260 remove((groupFile + toString(processIDS[i]) + ".temp").c_str());
261 for (int j = 0; j < fastaFileNames.size(); j++) {
262 appendFiles((fastaFileNames[j] + toString(processIDS[i]) + ".temp"), fastaFileNames[j]);
263 remove((fastaFileNames[j] + toString(processIDS[i]) + ".temp").c_str());
268 if (m->control_pressed) { return 0; }
272 openInputFile(fastaFile, inFASTA);
273 getNumSeqs(inFASTA, numSeqs);
276 lines.push_back(new linePair(0, numSeqs));
278 driverCreateTrim(fastaFile, qFileName, trimSeqFile, scrapSeqFile, groupFile, fastaFileNames, lines[0], lines[0]);
280 if (m->control_pressed) { return 0; }
284 for(int i=0;i<fastaFileNames.size();i++){
285 if (isBlank(fastaFileNames[i])) { remove(fastaFileNames[i].c_str()); }
289 //openInputFile(getRootName(fastaFile) + groupVector[i] + ".fasta", inFASTA);
290 openInputFile(fastaFileNames[i], inFASTA);
292 string outGroupFilename = outputDir + getRootName(getSimpleName(fastaFileNames[i])) + "groups";
293 openOutputFile(outGroupFilename, outGroups);
294 //openOutputFile(outputDir + getRootName(getSimpleName(fastaFile)) + groupVector[i] + ".groups", outGroups);
295 outputNames.push_back(outGroupFilename);
297 string thisGroup = "";
298 if (i > comboStarts) {
299 map<string, int>::iterator itCombo;
300 for(itCombo=combos.begin();itCombo!=combos.end(); itCombo++){
301 if(itCombo->second == i){ thisGroup = itCombo->first; combos.erase(itCombo); break; }
303 }else{ thisGroup = groupVector[i]; }
305 while(!inFASTA.eof()){
306 if(inFASTA.get() == '>'){
308 outGroups << seqName << '\t' << thisGroup << endl;
310 while (!inFASTA.eof()) { char c = inFASTA.get(); if (c == 10 || c == 13){ break; } }
317 if (m->control_pressed) {
318 for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); }
322 m->mothurOutEndLine();
323 m->mothurOut("Output File Names: "); m->mothurOutEndLine();
324 for (int i = 0; i < outputNames.size(); i++) { m->mothurOut(outputNames[i]); m->mothurOutEndLine(); }
325 m->mothurOutEndLine();
330 catch(exception& e) {
331 m->errorOut(e, "TrimSeqsCommand", "execute");
336 /**************************************************************************************/
337 int TrimSeqsCommand::driverCreateTrim(string filename, string qFileName, string trimFile, string scrapFile, string groupFile, vector<string> fastaNames, linePair* line, linePair* qline) {
341 int able = openOutputFile(trimFile, outFASTA);
344 openOutputFile(scrapFile, scrapFASTA);
347 vector<ofstream*> fastaFileNames;
348 if (oligoFile != "") {
349 openOutputFile(groupFile, outGroups);
350 for (int i = 0; i < fastaNames.size(); i++) {
351 #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
352 fastaFileNames.push_back(new ofstream((fastaNames[i] + toString(getpid()) + ".temp").c_str(), ios::ate));
354 fastaFileNames.push_back(new ofstream((fastaNames[i] + toString(i) + ".temp").c_str(), ios::ate));
360 openInputFile(filename, inFASTA);
363 if(qFileName != "") { openInputFile(qFileName, qFile); }
365 qFile.seekg(qline->start);
366 inFASTA.seekg(line->start);
368 for(int i=0;i<line->num;i++){
370 if (m->control_pressed) {
371 inFASTA.close(); outFASTA.close(); scrapFASTA.close(); if (oligoFile != "") { outGroups.close(); } if(qFileName != "") { qFile.close(); }
372 for(int i=0;i<fastaFileNames.size();i++){ fastaFileNames[i]->close(); delete fastaFileNames[i]; }
373 for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); }
379 Sequence currSeq(inFASTA);
380 cout << i << '\t' << currSeq.getName() << endl;
382 string origSeq = currSeq.getUnaligned();
384 int groupBar, groupPrime;
385 string trashCode = "";
386 int currentSeqsDiffs = 0;
389 if(qThreshold != 0) { success = stripQualThreshold(currSeq, qFile); }
390 else if(qAverage != 0) { success = cullQualAverage(currSeq, qFile); }
392 if (qtrim == 1 && (origSeq.length() != currSeq.getUnaligned().length())) {
393 success = 0; //if you don't want to trim and the sequence does not meet quality requirements, move to scrap
396 if(!success) { trashCode += 'q'; }
399 if(barcodes.size() != 0){
400 success = stripBarcode(currSeq, groupBar);
401 if(success > bdiffs){ trashCode += 'b'; }
402 else{ currentSeqsDiffs += success; }
405 if(numFPrimers != 0){
406 success = stripForward(currSeq, groupPrime);
407 if(success > pdiffs){ trashCode += 'f'; }
408 else{ currentSeqsDiffs += success; }
411 if (currentSeqsDiffs > tdiffs) { trashCode += 't'; }
413 if(numRPrimers != 0){
414 success = stripReverse(currSeq);
415 if(!success){ trashCode += 'r'; }
418 if(minLength > 0 || maxLength > 0){
419 success = cullLength(currSeq);
420 if(!success){ trashCode += 'l'; }
423 success = cullHomoP(currSeq);
424 if(!success){ trashCode += 'h'; }
427 success = cullAmbigs(currSeq);
428 if(!success){ trashCode += 'n'; }
431 if(flip){ currSeq.reverseComplement(); } // should go last
433 if(trashCode.length() == 0){
434 currSeq.setAligned(currSeq.getUnaligned());
435 currSeq.printSequence(outFASTA);
436 if(barcodes.size() != 0){
437 string thisGroup = groupVector[groupBar];
438 int indexToFastaFile = groupBar;
439 if (primers.size() != 0){
440 //does this primer have a group
441 if (groupVector[groupPrime] != "") {
442 thisGroup += "." + groupVector[groupPrime];
443 indexToFastaFile = combos[thisGroup];
446 outGroups << currSeq.getName() << '\t' << thisGroup << endl;
449 currSeq.printSequence(*fastaFileNames[indexToFastaFile]);
454 currSeq.setName(currSeq.getName() + '|' + trashCode);
455 currSeq.setUnaligned(origSeq);
456 currSeq.setAligned(origSeq);
457 currSeq.printSequence(scrapFASTA);
466 if (oligoFile != "") { outGroups.close(); }
467 if(qFileName != "") { qFile.close(); }
469 for(int i=0;i<fastaFileNames.size();i++){
470 fastaFileNames[i]->close();
471 delete fastaFileNames[i];
476 catch(exception& e) {
477 m->errorOut(e, "TrimSeqsCommand", "driverCreateTrim");
481 /**************************************************************************************************/
482 int TrimSeqsCommand::createProcessesCreateTrim(string filename, string qFileName, string trimFile, string scrapFile, string groupFile, vector<string> fastaNames) {
484 #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
489 //loop through and create all the processes you want
490 while (process != processors) {
494 processIDS.push_back(pid); //create map from line number to pid so you can append files in correct order later
497 driverCreateTrim(filename, qFileName, (trimFile + toString(getpid()) + ".temp"), (scrapFile + toString(getpid()) + ".temp"), (groupFile + toString(getpid()) + ".temp"), fastaNames, lines[process], qLines[process]);
499 }else { m->mothurOut("unable to spawn the necessary processes."); m->mothurOutEndLine(); exit(0); }
502 //force parent to wait until all the processes are done
503 for (int i=0;i<processors;i++) {
504 int temp = processIDS[i];
511 catch(exception& e) {
512 m->errorOut(e, "TrimSeqsCommand", "createProcessesCreateTrim");
516 /**************************************************************************************************/
518 int TrimSeqsCommand::setLines(string filename, vector<linePair*>& lines) {
523 vector<long int> positions;
526 openInputFile(filename, inFASTA);
529 while(!inFASTA.eof()){
530 input = getline(inFASTA);
532 if (input.length() != 0) {
533 if(input[0] == '>'){ long int pos = inFASTA.tellg(); positions.push_back(pos - input.length() - 1); }
538 int numFastaSeqs = positions.size();
543 //get num bytes in file
544 pFile = fopen (filename.c_str(),"rb");
545 if (pFile==NULL) perror ("Error opening file");
547 fseek (pFile, 0, SEEK_END);
552 int numSeqsPerProcessor = numFastaSeqs / processors;
554 for (int i = 0; i < processors; i++) {
556 long int startPos = positions[ i * numSeqsPerProcessor ];
557 if(i == processors - 1){
558 numSeqsPerProcessor = numFastaSeqs - i * numSeqsPerProcessor;
560 long int myEnd = positions[ (i+1) * numSeqsPerProcessor ];
562 lines.push_back(new linePair(startPos, numSeqsPerProcessor));
567 catch(exception& e) {
568 m->errorOut(e, "TrimSeqsCommand", "setLines");
572 //***************************************************************************************************************
574 void TrimSeqsCommand::getOligos(vector<string>& outFASTAVec){ //vector<ofstream*>& outFASTAVec
577 openInputFile(oligoFile, inOligos);
581 string type, oligo, group;
583 //int indexPrimer = 0;
585 while(!inOligos.eof()){
589 while (!inOligos.eof()) { char c = inOligos.get(); if (c == 10 || c == 13){ break; } } // get rest of line if there's any crap there
592 //make type case insensitive
593 for(int i=0;i<type.length();i++){ type[i] = toupper(type[i]); }
597 for(int i=0;i<oligo.length();i++){
598 oligo[i] = toupper(oligo[i]);
599 if(oligo[i] == 'U') { oligo[i] = 'T'; }
602 if(type == "FORWARD"){
605 // get rest of line in case there is a primer name
606 while (!inOligos.eof()) {
607 char c = inOligos.get();
608 if (c == 10 || c == 13){ break; }
609 else if (c == 32 || c == 9){;} //space or tab
613 //check for repeat barcodes
614 map<string, int>::iterator itPrime = primers.find(oligo);
615 if (itPrime != primers.end()) { m->mothurOut("primer " + oligo + " is in your oligos file already."); m->mothurOutEndLine(); }
617 primers[oligo]=index; index++;
618 groupVector.push_back(group);
621 if (group != "") { //there is a group for this primer
622 outputNames.push_back((outputDir + getRootName(getSimpleName(fastaFile)) + toString(index) + "." + group + ".fasta"));
623 outFASTAVec.push_back((outputDir + getRootName(getSimpleName(fastaFile)) + toString(index) + "." + group + ".fasta"));
628 else if(type == "REVERSE"){
629 Sequence oligoRC("reverse", oligo);
630 oligoRC.reverseComplement();
631 revPrimer.push_back(oligoRC.getUnaligned());
633 else if(type == "BARCODE"){
636 //check for repeat barcodes
637 map<string, int>::iterator itBar = barcodes.find(oligo);
638 if (itBar != barcodes.end()) { m->mothurOut("barcode " + oligo + " is in your oligos file already."); m->mothurOutEndLine(); }
640 barcodes[oligo]=index; index++;
641 groupVector.push_back(group);
644 outputNames.push_back((outputDir + getRootName(getSimpleName(fastaFile)) + toString(index) + "." + group + ".fasta"));
645 outFASTAVec.push_back((outputDir + getRootName(getSimpleName(fastaFile)) + toString(index) + "." + group + ".fasta"));
647 }else{ m->mothurOut(type + " is not recognized as a valid type. Choices are forward, reverse, and barcode. Ignoring " + oligo + "."); m->mothurOutEndLine(); }
654 //add in potential combos
656 comboStarts = outFASTAVec.size()-1;
657 for (map<string, int>::iterator itBar = barcodes.begin(); itBar != barcodes.end(); itBar++) {
658 for (map<string, int>::iterator itPrime = primers.begin(); itPrime != primers.end(); itPrime++) {
659 if (groupVector[itPrime->second] != "") { //there is a group for this primer
660 outputNames.push_back((outputDir + getRootName(getSimpleName(fastaFile)) + toString(itBar->second) + "." + groupVector[itBar->second] + "." + toString(itPrime->second) + "." + groupVector[itPrime->second] + ".fasta"));
661 outFASTAVec.push_back((outputDir + getRootName(getSimpleName(fastaFile)) + toString(itBar->second) + "." + groupVector[itBar->second] + "." + toString(itPrime->second) + "." + groupVector[itPrime->second] + ".fasta"));
662 combos[(groupVector[itBar->second] + "." + groupVector[itPrime->second])] = outFASTAVec.size()-1;
668 numFPrimers = primers.size();
669 numRPrimers = revPrimer.size();
672 catch(exception& e) {
673 m->errorOut(e, "TrimSeqsCommand", "getOligos");
677 //***************************************************************************************************************
679 int TrimSeqsCommand::stripBarcode(Sequence& seq, int& group){
682 string rawSequence = seq.getUnaligned();
683 int success = bdiffs + 1; //guilty until proven innocent
685 //can you find the barcode
686 for(map<string,int>::iterator it=barcodes.begin();it!=barcodes.end();it++){
687 string oligo = it->first;
688 if(rawSequence.length() < oligo.length()){ //let's just assume that the barcodes are the same length
689 success = bdiffs + 10; //if the sequence is shorter than the barcode then bail out
693 if(compareDNASeq(oligo, rawSequence.substr(0,oligo.length()))){
695 seq.setUnaligned(rawSequence.substr(oligo.length()));
701 //if you found the barcode or if you don't want to allow for diffs
703 if ((bdiffs == 0) || (success == 0)) { return success; }
705 else { //try aligning and see if you can find it
710 Alignment* alignment;
711 if (barcodes.size() > 0) {
712 map<string,int>::iterator it=barcodes.begin();
714 for(it;it!=barcodes.end();it++){
715 if(it->first.length() > maxLength){
716 maxLength = it->first.length();
719 alignment = new NeedlemanOverlap(-1.0, 1.0, -1.0, (maxLength+bdiffs+1));
721 }else{ alignment = NULL; }
723 //can you find the barcode
729 for(map<string,int>::iterator it=barcodes.begin();it!=barcodes.end();it++){
730 string oligo = it->first;
731 // int length = oligo.length();
733 if(rawSequence.length() < maxLength){ //let's just assume that the barcodes are the same length
734 success = bdiffs + 10;
738 //use needleman to align first barcode.length()+numdiffs of sequence to each barcode
739 alignment->align(oligo, rawSequence.substr(0,oligo.length()+bdiffs));
740 oligo = alignment->getSeqAAln();
741 string temp = alignment->getSeqBAln();
743 int alnLength = oligo.length();
745 for(int i=oligo.length()-1;i>=0;i--){
746 if(oligo[i] != '-'){ alnLength = i+1; break; }
748 oligo = oligo.substr(0,alnLength);
749 temp = temp.substr(0,alnLength);
752 int numDiff = countDiffs(oligo, temp);
754 // cout << oligo << '\t' << temp << '\t' << numDiff << endl;
756 if(numDiff < minDiff){
759 minGroup = it->second;
761 for(int i=0;i<alnLength;i++){
767 else if(numDiff == minDiff){
773 if(minDiff > bdiffs) { success = minDiff; } //no good matches
774 else if(minCount > 1) { success = bdiffs + 100; } //can't tell the difference between multiple barcodes
775 else{ //use the best match
777 seq.setUnaligned(rawSequence.substr(minPos));
781 if (alignment != NULL) { delete alignment; }
784 // cout << success << endl;
789 catch(exception& e) {
790 m->errorOut(e, "TrimSeqsCommand", "stripBarcode");
796 //***************************************************************************************************************
798 int TrimSeqsCommand::stripForward(Sequence& seq, int& group){
800 string rawSequence = seq.getUnaligned();
801 int success = pdiffs + 1; //guilty until proven innocent
803 //can you find the primer
804 for(map<string,int>::iterator it=primers.begin();it!=primers.end();it++){
805 string oligo = it->first;
806 if(rawSequence.length() < oligo.length()){ //let's just assume that the primers are the same length
807 success = pdiffs + 10; //if the sequence is shorter than the barcode then bail out
811 if(compareDNASeq(oligo, rawSequence.substr(0,oligo.length()))){
813 seq.setUnaligned(rawSequence.substr(oligo.length()));
819 //if you found the barcode or if you don't want to allow for diffs
821 if ((pdiffs == 0) || (success == 0)) { return success; }
823 else { //try aligning and see if you can find it
828 Alignment* alignment;
829 if (primers.size() > 0) {
830 map<string,int>::iterator it=primers.begin();
832 for(it;it!=primers.end();it++){
833 if(it->first.length() > maxLength){
834 maxLength = it->first.length();
837 alignment = new NeedlemanOverlap(-1.0, 1.0, -1.0, (maxLength+pdiffs+1));
839 }else{ alignment = NULL; }
841 //can you find the barcode
847 for(map<string,int>::iterator it=primers.begin();it!=primers.end();it++){
848 string oligo = it->first;
849 // int length = oligo.length();
851 if(rawSequence.length() < maxLength){
852 success = pdiffs + 100;
856 //use needleman to align first barcode.length()+numdiffs of sequence to each barcode
857 alignment->align(oligo, rawSequence.substr(0,oligo.length()+pdiffs));
858 oligo = alignment->getSeqAAln();
859 string temp = alignment->getSeqBAln();
861 int alnLength = oligo.length();
863 for(int i=oligo.length()-1;i>=0;i--){
864 if(oligo[i] != '-'){ alnLength = i+1; break; }
866 oligo = oligo.substr(0,alnLength);
867 temp = temp.substr(0,alnLength);
870 int numDiff = countDiffs(oligo, temp);
872 // cout << oligo << '\t' << temp << '\t' << numDiff << endl;
874 if(numDiff < minDiff){
877 minGroup = it->second;
879 for(int i=0;i<alnLength;i++){
885 else if(numDiff == minDiff){
891 if(minDiff > pdiffs) { success = minDiff; } //no good matches
892 else if(minCount > 1) { success = pdiffs + 10; } //can't tell the difference between multiple primers
893 else{ //use the best match
895 seq.setUnaligned(rawSequence.substr(minPos));
899 if (alignment != NULL) { delete alignment; }
906 catch(exception& e) {
907 m->errorOut(e, "TrimSeqsCommand", "stripForward");
912 //***************************************************************************************************************
914 bool TrimSeqsCommand::stripReverse(Sequence& seq){
916 string rawSequence = seq.getUnaligned();
917 bool success = 0; //guilty until proven innocent
919 for(int i=0;i<numRPrimers;i++){
920 string oligo = revPrimer[i];
922 if(rawSequence.length() < oligo.length()){
927 if(compareDNASeq(oligo, rawSequence.substr(rawSequence.length()-oligo.length(),oligo.length()))){
928 seq.setUnaligned(rawSequence.substr(0,rawSequence.length()-oligo.length()));
936 catch(exception& e) {
937 m->errorOut(e, "TrimSeqsCommand", "stripReverse");
942 //***************************************************************************************************************
944 bool TrimSeqsCommand::cullLength(Sequence& seq){
947 int length = seq.getNumBases();
948 bool success = 0; //guilty until proven innocent
950 if(length >= minLength && maxLength == 0) { success = 1; }
951 else if(length >= minLength && length <= maxLength) { success = 1; }
952 else { success = 0; }
957 catch(exception& e) {
958 m->errorOut(e, "TrimSeqsCommand", "cullLength");
964 //***************************************************************************************************************
966 bool TrimSeqsCommand::cullHomoP(Sequence& seq){
968 int longHomoP = seq.getLongHomoPolymer();
969 bool success = 0; //guilty until proven innocent
971 if(longHomoP <= maxHomoP){ success = 1; }
972 else { success = 0; }
976 catch(exception& e) {
977 m->errorOut(e, "TrimSeqsCommand", "cullHomoP");
983 //***************************************************************************************************************
985 bool TrimSeqsCommand::cullAmbigs(Sequence& seq){
987 int numNs = seq.getAmbigBases();
988 bool success = 0; //guilty until proven innocent
990 if(numNs <= maxAmbig) { success = 1; }
991 else { success = 0; }
995 catch(exception& e) {
996 m->errorOut(e, "TrimSeqsCommand", "cullAmbigs");
1002 //***************************************************************************************************************
1004 bool TrimSeqsCommand::compareDNASeq(string oligo, string seq){
1007 int length = oligo.length();
1009 for(int i=0;i<length;i++){
1011 if(oligo[i] != seq[i]){
1012 if(oligo[i] == 'A' || oligo[i] == 'T' || oligo[i] == 'G' || oligo[i] == 'C') { success = 0; }
1013 else if((oligo[i] == 'N' || oligo[i] == 'I') && (seq[i] == 'N')) { success = 0; }
1014 else if(oligo[i] == 'R' && (seq[i] != 'A' && seq[i] != 'G')) { success = 0; }
1015 else if(oligo[i] == 'Y' && (seq[i] != 'C' && seq[i] != 'T')) { success = 0; }
1016 else if(oligo[i] == 'M' && (seq[i] != 'C' && seq[i] != 'A')) { success = 0; }
1017 else if(oligo[i] == 'K' && (seq[i] != 'T' && seq[i] != 'G')) { success = 0; }
1018 else if(oligo[i] == 'W' && (seq[i] != 'T' && seq[i] != 'A')) { success = 0; }
1019 else if(oligo[i] == 'S' && (seq[i] != 'C' && seq[i] != 'G')) { success = 0; }
1020 else if(oligo[i] == 'B' && (seq[i] != 'C' && seq[i] != 'T' && seq[i] != 'G')) { success = 0; }
1021 else if(oligo[i] == 'D' && (seq[i] != 'A' && seq[i] != 'T' && seq[i] != 'G')) { success = 0; }
1022 else if(oligo[i] == 'H' && (seq[i] != 'A' && seq[i] != 'T' && seq[i] != 'C')) { success = 0; }
1023 else if(oligo[i] == 'V' && (seq[i] != 'A' && seq[i] != 'C' && seq[i] != 'G')) { success = 0; }
1025 if(success == 0) { break; }
1034 catch(exception& e) {
1035 m->errorOut(e, "TrimSeqsCommand", "compareDNASeq");
1040 //***************************************************************************************************************
1042 int TrimSeqsCommand::countDiffs(string oligo, string seq){
1045 int length = oligo.length();
1048 for(int i=0;i<length;i++){
1050 if(oligo[i] != seq[i]){
1051 if(oligo[i] == 'A' || oligo[i] == 'T' || oligo[i] == 'G' || oligo[i] == 'C' || oligo[i] == '-' || oligo[i] == '.') { countDiffs++; }
1052 else if((oligo[i] == 'N' || oligo[i] == 'I') && (seq[i] == 'N')) { countDiffs++; }
1053 else if(oligo[i] == 'R' && (seq[i] != 'A' && seq[i] != 'G')) { countDiffs++; }
1054 else if(oligo[i] == 'Y' && (seq[i] != 'C' && seq[i] != 'T')) { countDiffs++; }
1055 else if(oligo[i] == 'M' && (seq[i] != 'C' && seq[i] != 'A')) { countDiffs++; }
1056 else if(oligo[i] == 'K' && (seq[i] != 'T' && seq[i] != 'G')) { countDiffs++; }
1057 else if(oligo[i] == 'W' && (seq[i] != 'T' && seq[i] != 'A')) { countDiffs++; }
1058 else if(oligo[i] == 'S' && (seq[i] != 'C' && seq[i] != 'G')) { countDiffs++; }
1059 else if(oligo[i] == 'B' && (seq[i] != 'C' && seq[i] != 'T' && seq[i] != 'G')) { countDiffs++; }
1060 else if(oligo[i] == 'D' && (seq[i] != 'A' && seq[i] != 'T' && seq[i] != 'G')) { countDiffs++; }
1061 else if(oligo[i] == 'H' && (seq[i] != 'A' && seq[i] != 'T' && seq[i] != 'C')) { countDiffs++; }
1062 else if(oligo[i] == 'V' && (seq[i] != 'A' && seq[i] != 'C' && seq[i] != 'G')) { countDiffs++; }
1069 catch(exception& e) {
1070 m->errorOut(e, "TrimSeqsCommand", "countDiffs");
1075 //***************************************************************************************************************
1077 bool TrimSeqsCommand::stripQualThreshold(Sequence& seq, ifstream& qFile){
1079 // string rawSequence = seq.getUnaligned();
1080 // int seqLength; // = rawSequence.length();
1081 // string name, temp, temp2;
1085 // //get rest of line
1087 // while (!qFile.eof()) {
1088 // char c = qFile.get();
1089 // if (c == 10 || c == 13){ break; }
1090 // else { temp += c; }
1093 // int pos = temp.find("length");
1094 // if (pos == temp.npos) { m->mothurOut("Cannot find length in qfile for " + seq.getName()); m->mothurOutEndLine(); seqLength = 0; }
1096 // string tempLength = temp.substr(pos);
1097 // istringstream iss (tempLength,istringstream::in);
1101 // splitAtEquals(temp2, temp); //separates length=242, temp=length, temp2=242
1102 // convert(temp, seqLength); //converts string to int
1104 // if (name.length() != 0) { if(name.substr(1) != seq.getName()) { m->mothurOut("sequence name mismatch btwn fasta and qual file"); m->mothurOutEndLine(); } }
1106 string rawSequence = seq.getUnaligned();
1107 int seqLength = seq.getNumBases();
1108 bool success = 0; //guilty until proven innocent
1112 if (name[0] == '>') { if(name.substr(1) != seq.getName()) { m->mothurOut("sequence name mismatch btwn fasta: " + seq.getName() + " and qual file: " + name); m->mothurOutEndLine(); } }
1114 while (!qFile.eof()) { char c = qFile.get(); if (c == 10 || c == 13){ break; } }
1117 int end = seqLength;
1119 for(int i=0;i<seqLength;i++){
1122 if(score < qThreshold){
1127 for(int i=end+1;i<seqLength;i++){
1131 seq.setUnaligned(rawSequence.substr(0,end));
1135 catch(exception& e) {
1136 m->errorOut(e, "TrimSeqsCommand", "stripQualThreshold");
1141 //***************************************************************************************************************
1143 bool TrimSeqsCommand::cullQualAverage(Sequence& seq, ifstream& qFile){
1145 string rawSequence = seq.getUnaligned();
1146 int seqLength = seq.getNumBases();
1147 bool success = 0; //guilty until proven innocent
1151 if (name[0] == '>') { if(name.substr(1) != seq.getName()) { m->mothurOut("sequence name mismatch btwn fasta: " + seq.getName() + " and qual file: " + name); m->mothurOutEndLine(); } }
1153 while (!qFile.eof()) { char c = qFile.get(); if (c == 10 || c == 13){ break; } }
1158 for(int i=0;i<seqLength;i++){
1162 average /= seqLength;
1164 if(average >= qAverage) { success = 1; }
1165 else { success = 0; }
1169 catch(exception& e) {
1170 m->errorOut(e, "TrimSeqsCommand", "cullQualAverage");
1175 //***************************************************************************************************************