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", "qwindowaverage", "qstepsize", "qwindowsize", "qaverage", "rollaverage", "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
85 //check for optional parameter and set defaults
86 // ...at some point should added some additional type checking...
88 temp = validParameter.validFile(parameters, "flip", false);
89 if (temp == "not found"){ flip = 0; }
90 else if(isTrue(temp)) { flip = 1; }
92 temp = validParameter.validFile(parameters, "oligos", true);
93 if (temp == "not found"){ oligoFile = ""; }
94 else if(temp == "not open"){ abort = true; }
95 else { oligoFile = temp; }
97 temp = validParameter.validFile(parameters, "maxambig", false); if (temp == "not found") { temp = "-1"; }
98 convert(temp, maxAmbig);
100 temp = validParameter.validFile(parameters, "maxhomop", false); if (temp == "not found") { temp = "0"; }
101 convert(temp, maxHomoP);
103 temp = validParameter.validFile(parameters, "minlength", false); if (temp == "not found") { temp = "0"; }
104 convert(temp, minLength);
106 temp = validParameter.validFile(parameters, "maxlength", false); if (temp == "not found") { temp = "0"; }
107 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, "rollaverage", false); if (temp == "not found") { temp = "0"; }
132 convert(temp, qRollAverage);
134 temp = validParameter.validFile(parameters, "qwindowaverage", false);if (temp == "not found") { temp = "0"; }
135 convert(temp, qWindowAverage);
137 temp = validParameter.validFile(parameters, "qwindowsize", false); if (temp == "not found") { temp = "100"; }
138 convert(temp, qWindowSize);
140 temp = validParameter.validFile(parameters, "qstepsize", false); if (temp == "not found") { temp = "10"; }
141 convert(temp, qWindowStep);
143 temp = validParameter.validFile(parameters, "qaverage", false); if (temp == "not found") { temp = "0"; }
144 convert(temp, qAverage);
146 temp = validParameter.validFile(parameters, "allfiles", false); if (temp == "not found") { temp = "F"; }
147 allFiles = isTrue(temp);
149 temp = validParameter.validFile(parameters, "processors", false); if (temp == "not found") { temp = "1"; }
150 convert(temp, processors);
152 if(allFiles && oligoFile == ""){
153 m->mothurOut("You selected allfiles, but didn't enter an oligos file. Ignoring the allfiles request."); m->mothurOutEndLine();
155 if((qAverage != 0 && qThreshold != 0) && qFileName == ""){
156 m->mothurOut("You didn't provide a quality file name, quality criteria will be ignored."); m->mothurOutEndLine();
160 if(!flip && oligoFile=="" && !maxLength && !minLength && (maxAmbig==-1) && !maxHomoP && qFileName == ""){
161 m->mothurOut("You didn't set any options... quiting command."); m->mothurOutEndLine();
167 catch(exception& e) {
168 m->errorOut(e, "TrimSeqsCommand", "TrimSeqsCommand");
172 //**********************************************************************************************************************
174 void TrimSeqsCommand::help(){
176 m->mothurOut("The trim.seqs command reads a fastaFile and creates .....\n");
177 m->mothurOut("The trim.seqs command parameters are fasta, flip, oligos, maxambig, maxhomop, minlength, maxlength, qfile, qthreshold, qaverage, diffs, qtrim and allfiles.\n");
178 m->mothurOut("The fasta parameter is required.\n");
179 m->mothurOut("The flip parameter will output the reverse compliment of your trimmed sequence. The default is false.\n");
180 m->mothurOut("The oligos parameter .... The default is "".\n");
181 m->mothurOut("The maxambig parameter .... The default is -1.\n");
182 m->mothurOut("The maxhomop parameter .... The default is 0.\n");
183 m->mothurOut("The minlength parameter .... The default is 0.\n");
184 m->mothurOut("The maxlength parameter .... The default is 0.\n");
185 m->mothurOut("The tdiffs parameter is used to specify the total number of differences allowed in the sequence. The default is pdiffs + bdiffs.\n");
186 m->mothurOut("The bdiffs parameter is used to specify the number of differences allowed in the barcode. The default is 0.\n");
187 m->mothurOut("The pdiffs parameter is used to specify the number of differences allowed in the primer. The default is 0.\n");
188 m->mothurOut("The qfile parameter .....\n");
189 m->mothurOut("The qthreshold parameter .... The default is 0.\n");
190 m->mothurOut("The qaverage parameter .... The default is 0.\n");
191 m->mothurOut("The allfiles parameter .... The default is F.\n");
192 m->mothurOut("The qtrim parameter .... The default is F.\n");
193 m->mothurOut("The trim.seqs command should be in the following format: \n");
194 m->mothurOut("trim.seqs(fasta=yourFastaFile, flip=yourFlip, oligos=yourOligos, maxambig=yourMaxambig, \n");
195 m->mothurOut("maxhomop=yourMaxhomop, minlength=youMinlength, maxlength=yourMaxlength) \n");
196 m->mothurOut("Example trim.seqs(fasta=abrecovery.fasta, flip=..., oligos=..., maxambig=..., maxhomop=..., minlength=..., maxlength=...).\n");
197 m->mothurOut("Note: No spaces between parameter labels (i.e. fasta), '=' and parameters (i.e.yourFasta).\n");
198 m->mothurOut("For more details please check out the wiki http://www.mothur.org/wiki/Trim.seqs .\n\n");
201 catch(exception& e) {
202 m->errorOut(e, "TrimSeqsCommand", "help");
208 //***************************************************************************************************************
210 TrimSeqsCommand::~TrimSeqsCommand(){ /* do nothing */ }
212 //***************************************************************************************************************
214 int TrimSeqsCommand::execute(){
217 if (abort == true) { return 0; }
219 numFPrimers = 0; //this needs to be initialized
222 string trimSeqFile = outputDir + getRootName(getSimpleName(fastaFile)) + "trim.fasta";
223 outputNames.push_back(trimSeqFile);
224 string scrapSeqFile = outputDir + getRootName(getSimpleName(fastaFile)) + "scrap.fasta";
225 outputNames.push_back(scrapSeqFile);
226 string trimQualFile = outputDir + getRootName(getSimpleName(fastaFile)) + "trim.qual";
227 outputNames.push_back(trimQualFile);
228 string scrapQualFile = outputDir + getRootName(getSimpleName(fastaFile)) + "scrap.qual";
229 outputNames.push_back(scrapQualFile);
230 string groupFile = outputDir + getRootName(getSimpleName(fastaFile)) + "groups";
232 vector<string> fastaFileNames;
233 vector<string> qualFileNames;
235 outputNames.push_back(groupFile);
236 getOligos(fastaFileNames, qualFileNames);
239 if(qFileName != "") { setLines(qFileName, qLines); }
242 #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
246 openInputFile(fastaFile, inFASTA);
247 getNumSeqs(inFASTA, numSeqs);
250 lines.push_back(new linePair(0, numSeqs));
252 driverCreateTrim(fastaFile, qFileName, trimSeqFile, scrapSeqFile, trimQualFile, scrapQualFile, groupFile, fastaFileNames, qualFileNames, lines[0], lines[0]);
254 for (int j = 0; j < fastaFileNames.size(); j++) {
255 rename((fastaFileNames[j] + toString(getpid()) + ".temp").c_str(), fastaFileNames[j].c_str());
258 for (int j = 0; j < qualFileNames.size(); j++) {
259 rename((qualFileNames[j] + toString(getpid()) + ".temp").c_str(), qualFileNames[j].c_str());
264 setLines(fastaFile, lines);
265 if(qFileName == "") { qLines = lines; }
267 createProcessesCreateTrim(fastaFile, qFileName, trimSeqFile, scrapSeqFile, trimQualFile, scrapQualFile, groupFile, fastaFileNames, qualFileNames);
269 rename((trimSeqFile + toString(processIDS[0]) + ".temp").c_str(), trimSeqFile.c_str());
270 rename((scrapSeqFile + toString(processIDS[0]) + ".temp").c_str(), scrapSeqFile.c_str());
271 rename((groupFile + toString(processIDS[0]) + ".temp").c_str(), groupFile.c_str());
274 rename((trimQualFile + toString(processIDS[0]) + ".temp").c_str(), trimQualFile.c_str());
275 rename((scrapQualFile + toString(processIDS[0]) + ".temp").c_str(), scrapQualFile.c_str());
279 for (int j = 0; j < fastaFileNames.size(); j++) {
280 rename((fastaFileNames[j] + toString(processIDS[0]) + ".temp").c_str(), fastaFileNames[j].c_str());
283 for (int j = 0; j < qualFileNames.size(); j++) {
284 rename((qualFileNames[j] + toString(getpid()) + ".temp").c_str(), qualFileNames[j].c_str());
289 for(int i=1;i<processors;i++){
290 appendFiles((trimSeqFile + toString(processIDS[i]) + ".temp"), trimSeqFile);
291 remove((trimSeqFile + toString(processIDS[i]) + ".temp").c_str());
292 appendFiles((scrapSeqFile + toString(processIDS[i]) + ".temp"), scrapSeqFile);
293 remove((scrapSeqFile + toString(processIDS[i]) + ".temp").c_str());
295 appendFiles((trimQualFile + toString(processIDS[i]) + ".temp"), trimQualFile);
296 remove((trimQualFile + toString(processIDS[i]) + ".temp").c_str());
297 appendFiles((scrapQualFile + toString(processIDS[i]) + ".temp"), scrapQualFile);
298 remove((scrapQualFile + toString(processIDS[i]) + ".temp").c_str());
300 appendFiles((groupFile + toString(processIDS[i]) + ".temp"), groupFile);
301 remove((groupFile + toString(processIDS[i]) + ".temp").c_str());
302 for (int j = 0; j < fastaFileNames.size(); j++) {
303 appendFiles((fastaFileNames[j] + toString(processIDS[i]) + ".temp"), fastaFileNames[j]);
304 remove((fastaFileNames[j] + toString(processIDS[i]) + ".temp").c_str());
308 for (int j = 0; j < qualFileNames.size(); j++) {
309 appendFiles((qualFileNames[j] + toString(processIDS[i]) + ".temp"), qualFileNames[j]);
310 remove((qualFileNames[j] + toString(processIDS[i]) + ".temp").c_str());
318 if (m->control_pressed) { return 0; }
322 openInputFile(fastaFile, inFASTA);
323 getNumSeqs(inFASTA, numSeqs);
326 lines.push_back(new linePair(0, numSeqs));
328 driverCreateTrim(fastaFile, qFileName, trimSeqFile, scrapSeqFile, trimQualFile, scrapQualFile, groupFile, fastaFileNames, qualFileNames, lines[0], lines[0]);
330 if (m->control_pressed) { return 0; }
334 for(int i=0;i<fastaFileNames.size();i++){
335 if (isBlank(fastaFileNames[i])) { remove(fastaFileNames[i].c_str()); }
336 else if (filesToRemove.count(fastaFileNames[i]) > 0) { remove(fastaFileNames[i].c_str()); }
340 openInputFile(fastaFileNames[i], inFASTA);
342 string outGroupFilename = outputDir + getRootName(getSimpleName(fastaFileNames[i])) + "groups";
343 openOutputFile(outGroupFilename, outGroups);
344 outputNames.push_back(outGroupFilename);
346 string thisGroup = "";
347 if (i > comboStarts) {
348 map<string, int>::iterator itCombo;
349 for(itCombo=combos.begin();itCombo!=combos.end(); itCombo++){
350 if(itCombo->second == i){ thisGroup = itCombo->first; combos.erase(itCombo); break; }
352 }else{ thisGroup = groupVector[i]; }
354 while(!inFASTA.eof()){
355 if(inFASTA.get() == '>'){
357 outGroups << seqName << '\t' << thisGroup << endl;
359 while (!inFASTA.eof()) { char c = inFASTA.get(); if (c == 10 || c == 13){ break; } }
367 for(int i=0;i<qualFileNames.size();i++){
368 if (isBlank(qualFileNames[i])) { remove(qualFileNames[i].c_str()); }
369 else if (filesToRemove.count(qualFileNames[i]) > 0) { remove(qualFileNames[i].c_str()); }
373 openInputFile(qualFileNames[i], inQual);
374 // ofstream outGroups;
376 // string thisGroup = "";
377 // if (i > comboStarts) {
378 // map<string, int>::iterator itCombo;
379 // for(itCombo=combos.begin();itCombo!=combos.end(); itCombo++){
380 // if(itCombo->second == i){ thisGroup = itCombo->first; combos.erase(itCombo); break; }
383 // else{ thisGroup = groupVector[i]; }
391 if (m->control_pressed) {
392 for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); }
396 m->mothurOutEndLine();
397 m->mothurOut("Output File Names: "); m->mothurOutEndLine();
398 for (int i = 0; i < outputNames.size(); i++) { m->mothurOut(outputNames[i]); m->mothurOutEndLine(); }
399 m->mothurOutEndLine();
404 catch(exception& e) {
405 m->errorOut(e, "TrimSeqsCommand", "execute");
410 /**************************************************************************************/
412 int TrimSeqsCommand::driverCreateTrim(string filename, string qFileName, string trimFile, string scrapFile, string trimQFile, string scrapQFile, string groupFile, vector<string> fastaNames, vector<string> qualNames, linePair* line, linePair* qline) {
417 int able = openOutputFile(trimFile, outFASTA);
420 openOutputFile(scrapFile, scrapFASTA);
425 openOutputFile(trimQFile, outQual);
426 openOutputFile(scrapQFile, scrapQual);
430 vector<ofstream*> fastaFileNames;
431 vector<ofstream*> qualFileNames;
434 if (oligoFile != "") {
435 openOutputFile(groupFile, outGroups);
436 for (int i = 0; i < fastaNames.size(); i++) {
437 #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
438 fastaFileNames.push_back(new ofstream((fastaNames[i] + toString(getpid()) + ".temp").c_str(), ios::ate));
440 qualFileNames.push_back(new ofstream((qualNames[i] + toString(getpid()) + ".temp").c_str(), ios::ate));
443 fastaFileNames.push_back(new ofstream((fastaNames[i] + toString(i) + ".temp").c_str(), ios::ate));
445 qualFileNames.push_back(new ofstream((qualNames[i] + toString(i) + ".temp").c_str(), ios::ate));
452 openInputFile(filename, inFASTA);
455 if(qFileName != "") { openInputFile(qFileName, qFile); }
457 qFile.seekg(qline->start);
458 inFASTA.seekg(line->start);
460 for(int i=0;i<line->num;i++){
462 if (m->control_pressed) {
463 inFASTA.close(); outFASTA.close(); scrapFASTA.close();
464 if (oligoFile != "") { outGroups.close(); }
466 for(int i=0;i<fastaFileNames.size();i++){ fastaFileNames[i]->close(); delete fastaFileNames[i]; }
470 for(int i=0;i<qualFileNames.size();i++){ qualFileNames[i]->close(); delete qualFileNames[i]; }
472 for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); }
479 Sequence currSeq(inFASTA);
480 QualityScores currQual;
482 currQual = QualityScores(qFile);
485 string origSeq = currSeq.getUnaligned();
487 int groupBar, groupPrime;
488 string trashCode = "";
489 int currentSeqsDiffs = 0;
492 if(qThreshold != 0) { success = currQual.stripQualThreshold(currSeq, qThreshold); }
493 else if(qAverage != 0) { success = currQual.cullQualAverage(currSeq, qAverage); }
494 else if(qRollAverage != 0) { success = currQual.stripQualRollingAverage(currSeq, qRollAverage); }
495 else if(qWindowAverage != 0){ success = currQual.stripQualWindowAverage(currSeq, qWindowStep, qWindowSize, qWindowAverage); }
497 if (qtrim == 1 && (origSeq.length() != currSeq.getUnaligned().length())) {
498 success = 0; //if you don't want to trim and the sequence does not meet quality requirements, move to scrap
501 if(!success) { trashCode += 'q'; }
504 if(barcodes.size() != 0){
505 success = stripBarcode(currSeq, currQual, groupBar);
506 if(success > bdiffs) { trashCode += 'b'; }
507 else{ currentSeqsDiffs += success; }
510 if(numFPrimers != 0){
511 success = stripForward(currSeq, currQual, groupPrime);
512 if(success > pdiffs) { trashCode += 'f'; }
513 else{ currentSeqsDiffs += success; }
516 if (currentSeqsDiffs > tdiffs) { trashCode += 't'; }
518 if(numRPrimers != 0){
519 success = stripReverse(currSeq, currQual);
520 if(!success) { trashCode += 'r'; }
523 if(minLength > 0 || maxLength > 0){
524 success = cullLength(currSeq);
525 if(!success) { trashCode += 'l'; }
528 success = cullHomoP(currSeq);
529 if(!success) { trashCode += 'h'; }
532 success = cullAmbigs(currSeq);
533 if(!success) { trashCode += 'n'; }
536 if(flip){ currSeq.reverseComplement(); currQual.flipQScores(); } // should go last
538 if(trashCode.length() == 0){
539 currSeq.setAligned(currSeq.getUnaligned());
540 currSeq.printSequence(outFASTA);
541 currQual.printQScores(outQual);
543 if(barcodes.size() != 0){
544 string thisGroup = groupVector[groupBar];
545 int indexToFastaFile = groupBar;
546 if (primers.size() != 0){
547 //does this primer have a group
548 if (groupVector[groupPrime] != "") {
549 thisGroup += "." + groupVector[groupPrime];
550 indexToFastaFile = combos[thisGroup];
553 outGroups << currSeq.getName() << '\t' << thisGroup << endl;
555 currSeq.printSequence(*fastaFileNames[indexToFastaFile]);
558 currQual.printQScores(*qualFileNames[indexToFastaFile]);
564 currSeq.setName(currSeq.getName() + '|' + trashCode);
565 currSeq.setUnaligned(origSeq);
566 currSeq.setAligned(origSeq);
567 currSeq.printSequence(scrapFASTA);
568 currQual.printQScores(scrapQual);
578 if (oligoFile != "") { outGroups.close(); }
579 if(qFileName != "") { qFile.close(); scrapQual.close(); outQual.close(); }
581 for(int i=0;i<fastaFileNames.size();i++){
582 fastaFileNames[i]->close();
583 delete fastaFileNames[i];
587 for(int i=0;i<qualFileNames.size();i++){
588 qualFileNames[i]->close();
589 delete qualFileNames[i];
595 catch(exception& e) {
596 m->errorOut(e, "TrimSeqsCommand", "driverCreateTrim");
601 /**************************************************************************************************/
603 int TrimSeqsCommand::createProcessesCreateTrim(string filename, string qFileName, string trimFile, string scrapFile, string trimQFile, string scrapQFile, string groupFile, vector<string> fastaNames, vector<string> qualNames) {
605 #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
610 //loop through and create all the processes you want
611 while (process != processors) {
615 processIDS.push_back(pid); //create map from line number to pid so you can append files in correct order later
618 driverCreateTrim(filename, qFileName, (trimFile + toString(getpid()) + ".temp"), (scrapFile + toString(getpid()) + ".temp"), (trimQFile + toString(getpid()) + ".temp"), (scrapQFile + toString(getpid()) + ".temp"), (groupFile + toString(getpid()) + ".temp"), fastaNames, qualNames, lines[process], qLines[process]);
620 }else { m->mothurOut("unable to spawn the necessary processes."); m->mothurOutEndLine(); exit(0); }
623 //force parent to wait until all the processes are done
624 for (int i=0;i<processors;i++) {
625 int temp = processIDS[i];
632 catch(exception& e) {
633 m->errorOut(e, "TrimSeqsCommand", "createProcessesCreateTrim");
638 /**************************************************************************************************/
640 int TrimSeqsCommand::setLines(string filename, vector<linePair*>& lines) {
645 vector<unsigned long int> positions;
648 openInputFile(filename, inFASTA);
651 while(!inFASTA.eof()){
652 input = getline(inFASTA);
654 if (input.length() != 0) {
655 if(input[0] == '>'){ unsigned long int pos = inFASTA.tellg(); positions.push_back(pos - input.length() - 1); }
660 int numFastaSeqs = positions.size();
663 unsigned long int size;
665 //get num bytes in file
666 pFile = fopen (filename.c_str(),"rb");
667 if (pFile==NULL) perror ("Error opening file");
669 fseek (pFile, 0, SEEK_END);
674 int numSeqsPerProcessor = numFastaSeqs / processors;
676 for (int i = 0; i < processors; i++) {
678 unsigned long int startPos = positions[ i * numSeqsPerProcessor ];
679 if(i == processors - 1){
680 numSeqsPerProcessor = numFastaSeqs - i * numSeqsPerProcessor;
682 unsigned long int myEnd = positions[ (i+1) * numSeqsPerProcessor ];
684 lines.push_back(new linePair(startPos, numSeqsPerProcessor));
689 catch(exception& e) {
690 m->errorOut(e, "TrimSeqsCommand", "setLines");
694 //***************************************************************************************************************
696 void TrimSeqsCommand::getOligos(vector<string>& outFASTAVec, vector<string>& outQualVec){
699 openInputFile(oligoFile, inOligos);
703 string type, oligo, group;
705 //int indexPrimer = 0;
707 while(!inOligos.eof()){
711 while (!inOligos.eof()) { char c = inOligos.get(); if (c == 10 || c == 13){ break; } } // get rest of line if there's any crap there
714 //make type case insensitive
715 for(int i=0;i<type.length();i++){ type[i] = toupper(type[i]); }
719 for(int i=0;i<oligo.length();i++){
720 oligo[i] = toupper(oligo[i]);
721 if(oligo[i] == 'U') { oligo[i] = 'T'; }
724 if(type == "FORWARD"){
727 // get rest of line in case there is a primer name
728 while (!inOligos.eof()) {
729 char c = inOligos.get();
730 if (c == 10 || c == 13){ break; }
731 else if (c == 32 || c == 9){;} //space or tab
735 //check for repeat barcodes
736 map<string, int>::iterator itPrime = primers.find(oligo);
737 if (itPrime != primers.end()) { m->mothurOut("primer " + oligo + " is in your oligos file already."); m->mothurOutEndLine(); }
739 primers[oligo]=index; index++;
740 groupVector.push_back(group);
743 outFASTAVec.push_back((outputDir + getRootName(getSimpleName(fastaFile)) + group + ".fasta"));
745 outQualVec.push_back((outputDir + getRootName(getSimpleName(qFileName)) + group + ".qual"));
747 if (group == "") { //if there is not a group for this primer, then this file will not get written to, but we add it to keep the indexes correct
748 filesToRemove.insert((outputDir + getRootName(getSimpleName(fastaFile)) + group + ".fasta"));
750 filesToRemove.insert((outputDir + getRootName(getSimpleName(qFileName)) + group + ".qual"));
753 outputNames.push_back((outputDir + getRootName(getSimpleName(fastaFile)) + group + ".fasta"));
755 outputNames.push_back((outputDir + getRootName(getSimpleName(qFileName)) + group + ".qual"));
761 else if(type == "REVERSE"){
762 Sequence oligoRC("reverse", oligo);
763 oligoRC.reverseComplement();
764 revPrimer.push_back(oligoRC.getUnaligned());
766 else if(type == "BARCODE"){
769 //check for repeat barcodes
770 map<string, int>::iterator itBar = barcodes.find(oligo);
771 if (itBar != barcodes.end()) { m->mothurOut("barcode " + oligo + " is in your oligos file already."); m->mothurOutEndLine(); }
773 barcodes[oligo]=index; index++;
774 groupVector.push_back(group);
777 outputNames.push_back((outputDir + getRootName(getSimpleName(fastaFile)) + group + ".fasta"));
778 outFASTAVec.push_back((outputDir + getRootName(getSimpleName(fastaFile)) + group + ".fasta"));
780 outQualVec.push_back((outputDir + getRootName(getSimpleName(qFileName)) + group + ".qual"));
781 outputNames.push_back((outputDir + getRootName(getSimpleName(qFileName)) + group + ".qual"));
784 }else{ m->mothurOut(type + " is not recognized as a valid type. Choices are forward, reverse, and barcode. Ignoring " + oligo + "."); m->mothurOutEndLine(); }
791 //add in potential combos
793 comboStarts = outFASTAVec.size()-1;
794 for (map<string, int>::iterator itBar = barcodes.begin(); itBar != barcodes.end(); itBar++) {
795 for (map<string, int>::iterator itPrime = primers.begin(); itPrime != primers.end(); itPrime++) {
796 if (groupVector[itPrime->second] != "") { //there is a group for this primer
797 outputNames.push_back((outputDir + getRootName(getSimpleName(qFileName)) + groupVector[itBar->second] + "." + groupVector[itPrime->second] + ".fasta"));
798 outFASTAVec.push_back((outputDir + getRootName(getSimpleName(fastaFile)) + groupVector[itBar->second] + "." + groupVector[itPrime->second] + ".fasta"));
799 combos[(groupVector[itBar->second] + "." + groupVector[itPrime->second])] = outFASTAVec.size()-1;
802 outQualVec.push_back((outputDir + getRootName(getSimpleName(qFileName)) + groupVector[itBar->second] + "." + groupVector[itPrime->second] + ".qual"));
803 outputNames.push_back((outputDir + getRootName(getSimpleName(qFileName)) + groupVector[itBar->second] + "." + groupVector[itPrime->second] + ".qual"));
810 numFPrimers = primers.size();
811 numRPrimers = revPrimer.size();
814 catch(exception& e) {
815 m->errorOut(e, "TrimSeqsCommand", "getOligos");
819 //***************************************************************************************************************
821 int TrimSeqsCommand::stripBarcode(Sequence& seq, QualityScores& qual, int& group){
824 string rawSequence = seq.getUnaligned();
825 int success = bdiffs + 1; //guilty until proven innocent
827 //can you find the barcode
828 for(map<string,int>::iterator it=barcodes.begin();it!=barcodes.end();it++){
829 string oligo = it->first;
830 if(rawSequence.length() < oligo.length()){ //let's just assume that the barcodes are the same length
831 success = bdiffs + 10; //if the sequence is shorter than the barcode then bail out
835 if(compareDNASeq(oligo, rawSequence.substr(0,oligo.length()))){
837 seq.setUnaligned(rawSequence.substr(oligo.length()));
839 if(qual.getName() != ""){
840 qual.trimQScores(oligo.length(), -1);
848 //if you found the barcode or if you don't want to allow for diffs
850 if ((bdiffs == 0) || (success == 0)) { return success; }
852 else { //try aligning and see if you can find it
857 Alignment* alignment;
858 if (barcodes.size() > 0) {
859 map<string,int>::iterator it=barcodes.begin();
861 for(it;it!=barcodes.end();it++){
862 if(it->first.length() > maxLength){
863 maxLength = it->first.length();
866 alignment = new NeedlemanOverlap(-1.0, 1.0, -1.0, (maxLength+bdiffs+1));
868 }else{ alignment = NULL; }
870 //can you find the barcode
876 for(map<string,int>::iterator it=barcodes.begin();it!=barcodes.end();it++){
877 string oligo = it->first;
878 // int length = oligo.length();
880 if(rawSequence.length() < maxLength){ //let's just assume that the barcodes are the same length
881 success = bdiffs + 10;
885 //use needleman to align first barcode.length()+numdiffs of sequence to each barcode
886 alignment->align(oligo, rawSequence.substr(0,oligo.length()+bdiffs));
887 oligo = alignment->getSeqAAln();
888 string temp = alignment->getSeqBAln();
890 int alnLength = oligo.length();
892 for(int i=oligo.length()-1;i>=0;i--){
893 if(oligo[i] != '-'){ alnLength = i+1; break; }
895 oligo = oligo.substr(0,alnLength);
896 temp = temp.substr(0,alnLength);
899 int numDiff = countDiffs(oligo, temp);
901 // cout << oligo << '\t' << temp << '\t' << numDiff << endl;
903 if(numDiff < minDiff){
906 minGroup = it->second;
908 for(int i=0;i<alnLength;i++){
914 else if(numDiff == minDiff){
920 if(minDiff > bdiffs) { success = minDiff; } //no good matches
921 else if(minCount > 1) { success = bdiffs + 100; } //can't tell the difference between multiple barcodes
922 else{ //use the best match
924 seq.setUnaligned(rawSequence.substr(minPos));
926 if(qual.getName() != ""){
927 qual.trimQScores(minPos, -1);
932 if (alignment != NULL) { delete alignment; }
935 // cout << success << endl;
940 catch(exception& e) {
941 m->errorOut(e, "TrimSeqsCommand", "stripBarcode");
947 //***************************************************************************************************************
949 int TrimSeqsCommand::stripForward(Sequence& seq, QualityScores& qual, int& group){
951 string rawSequence = seq.getUnaligned();
952 int success = pdiffs + 1; //guilty until proven innocent
954 //can you find the primer
955 for(map<string,int>::iterator it=primers.begin();it!=primers.end();it++){
956 string oligo = it->first;
957 if(rawSequence.length() < oligo.length()){ //let's just assume that the primers are the same length
958 success = pdiffs + 10; //if the sequence is shorter than the barcode then bail out
962 if(compareDNASeq(oligo, rawSequence.substr(0,oligo.length()))){
964 seq.setUnaligned(rawSequence.substr(oligo.length()));
965 if(qual.getName() != ""){
966 qual.trimQScores(oligo.length(), -1);
974 //if you found the barcode or if you don't want to allow for diffs
976 if ((pdiffs == 0) || (success == 0)) { return success; }
978 else { //try aligning and see if you can find it
983 Alignment* alignment;
984 if (primers.size() > 0) {
985 map<string,int>::iterator it=primers.begin();
987 for(it;it!=primers.end();it++){
988 if(it->first.length() > maxLength){
989 maxLength = it->first.length();
992 alignment = new NeedlemanOverlap(-1.0, 1.0, -1.0, (maxLength+pdiffs+1));
994 }else{ alignment = NULL; }
996 //can you find the barcode
1002 for(map<string,int>::iterator it=primers.begin();it!=primers.end();it++){
1003 string oligo = it->first;
1004 // int length = oligo.length();
1006 if(rawSequence.length() < maxLength){
1007 success = pdiffs + 100;
1011 //use needleman to align first barcode.length()+numdiffs of sequence to each barcode
1012 alignment->align(oligo, rawSequence.substr(0,oligo.length()+pdiffs));
1013 oligo = alignment->getSeqAAln();
1014 string temp = alignment->getSeqBAln();
1016 int alnLength = oligo.length();
1018 for(int i=oligo.length()-1;i>=0;i--){
1019 if(oligo[i] != '-'){ alnLength = i+1; break; }
1021 oligo = oligo.substr(0,alnLength);
1022 temp = temp.substr(0,alnLength);
1025 int numDiff = countDiffs(oligo, temp);
1027 // cout << oligo << '\t' << temp << '\t' << numDiff << endl;
1029 if(numDiff < minDiff){
1032 minGroup = it->second;
1034 for(int i=0;i<alnLength;i++){
1040 else if(numDiff == minDiff){
1046 if(minDiff > pdiffs) { success = minDiff; } //no good matches
1047 else if(minCount > 1) { success = pdiffs + 10; } //can't tell the difference between multiple primers
1048 else{ //use the best match
1050 seq.setUnaligned(rawSequence.substr(minPos));
1051 if(qual.getName() != ""){
1052 qual.trimQScores(minPos, -1);
1057 if (alignment != NULL) { delete alignment; }
1064 catch(exception& e) {
1065 m->errorOut(e, "TrimSeqsCommand", "stripForward");
1070 //***************************************************************************************************************
1072 bool TrimSeqsCommand::stripReverse(Sequence& seq, QualityScores& qual){
1074 string rawSequence = seq.getUnaligned();
1075 bool success = 0; //guilty until proven innocent
1077 for(int i=0;i<numRPrimers;i++){
1078 string oligo = revPrimer[i];
1080 if(rawSequence.length() < oligo.length()){
1085 if(compareDNASeq(oligo, rawSequence.substr(rawSequence.length()-oligo.length(),oligo.length()))){
1086 seq.setUnaligned(rawSequence.substr(0,rawSequence.length()-oligo.length()));
1087 if(qual.getName() != ""){
1088 qual.trimQScores(-1, rawSequence.length()-oligo.length());
1097 catch(exception& e) {
1098 m->errorOut(e, "TrimSeqsCommand", "stripReverse");
1103 //***************************************************************************************************************
1105 bool TrimSeqsCommand::cullLength(Sequence& seq){
1108 int length = seq.getNumBases();
1109 bool success = 0; //guilty until proven innocent
1111 if(length >= minLength && maxLength == 0) { success = 1; }
1112 else if(length >= minLength && length <= maxLength) { success = 1; }
1113 else { success = 0; }
1118 catch(exception& e) {
1119 m->errorOut(e, "TrimSeqsCommand", "cullLength");
1125 //***************************************************************************************************************
1127 bool TrimSeqsCommand::cullHomoP(Sequence& seq){
1129 int longHomoP = seq.getLongHomoPolymer();
1130 bool success = 0; //guilty until proven innocent
1132 if(longHomoP <= maxHomoP){ success = 1; }
1133 else { success = 0; }
1137 catch(exception& e) {
1138 m->errorOut(e, "TrimSeqsCommand", "cullHomoP");
1144 //***************************************************************************************************************
1146 bool TrimSeqsCommand::cullAmbigs(Sequence& seq){
1148 int numNs = seq.getAmbigBases();
1149 bool success = 0; //guilty until proven innocent
1151 if(numNs <= maxAmbig) { success = 1; }
1152 else { success = 0; }
1156 catch(exception& e) {
1157 m->errorOut(e, "TrimSeqsCommand", "cullAmbigs");
1163 //***************************************************************************************************************
1165 bool TrimSeqsCommand::compareDNASeq(string oligo, string seq){
1168 int length = oligo.length();
1170 for(int i=0;i<length;i++){
1172 if(oligo[i] != seq[i]){
1173 if(oligo[i] == 'A' || oligo[i] == 'T' || oligo[i] == 'G' || oligo[i] == 'C') { success = 0; }
1174 else if((oligo[i] == 'N' || oligo[i] == 'I') && (seq[i] == 'N')) { success = 0; }
1175 else if(oligo[i] == 'R' && (seq[i] != 'A' && seq[i] != 'G')) { success = 0; }
1176 else if(oligo[i] == 'Y' && (seq[i] != 'C' && seq[i] != 'T')) { success = 0; }
1177 else if(oligo[i] == 'M' && (seq[i] != 'C' && seq[i] != 'A')) { success = 0; }
1178 else if(oligo[i] == 'K' && (seq[i] != 'T' && seq[i] != 'G')) { success = 0; }
1179 else if(oligo[i] == 'W' && (seq[i] != 'T' && seq[i] != 'A')) { success = 0; }
1180 else if(oligo[i] == 'S' && (seq[i] != 'C' && seq[i] != 'G')) { success = 0; }
1181 else if(oligo[i] == 'B' && (seq[i] != 'C' && seq[i] != 'T' && seq[i] != 'G')) { success = 0; }
1182 else if(oligo[i] == 'D' && (seq[i] != 'A' && seq[i] != 'T' && seq[i] != 'G')) { success = 0; }
1183 else if(oligo[i] == 'H' && (seq[i] != 'A' && seq[i] != 'T' && seq[i] != 'C')) { success = 0; }
1184 else if(oligo[i] == 'V' && (seq[i] != 'A' && seq[i] != 'C' && seq[i] != 'G')) { success = 0; }
1186 if(success == 0) { break; }
1195 catch(exception& e) {
1196 m->errorOut(e, "TrimSeqsCommand", "compareDNASeq");
1201 //***************************************************************************************************************
1203 int TrimSeqsCommand::countDiffs(string oligo, string seq){
1206 int length = oligo.length();
1209 for(int i=0;i<length;i++){
1211 if(oligo[i] != seq[i]){
1212 if(oligo[i] == 'A' || oligo[i] == 'T' || oligo[i] == 'G' || oligo[i] == 'C' || oligo[i] == '-' || oligo[i] == '.') { countDiffs++; }
1213 else if((oligo[i] == 'N' || oligo[i] == 'I') && (seq[i] == 'N')) { countDiffs++; }
1214 else if(oligo[i] == 'R' && (seq[i] != 'A' && seq[i] != 'G')) { countDiffs++; }
1215 else if(oligo[i] == 'Y' && (seq[i] != 'C' && seq[i] != 'T')) { countDiffs++; }
1216 else if(oligo[i] == 'M' && (seq[i] != 'C' && seq[i] != 'A')) { countDiffs++; }
1217 else if(oligo[i] == 'K' && (seq[i] != 'T' && seq[i] != 'G')) { countDiffs++; }
1218 else if(oligo[i] == 'W' && (seq[i] != 'T' && seq[i] != 'A')) { countDiffs++; }
1219 else if(oligo[i] == 'S' && (seq[i] != 'C' && seq[i] != 'G')) { countDiffs++; }
1220 else if(oligo[i] == 'B' && (seq[i] != 'C' && seq[i] != 'T' && seq[i] != 'G')) { countDiffs++; }
1221 else if(oligo[i] == 'D' && (seq[i] != 'A' && seq[i] != 'T' && seq[i] != 'G')) { countDiffs++; }
1222 else if(oligo[i] == 'H' && (seq[i] != 'A' && seq[i] != 'T' && seq[i] != 'C')) { countDiffs++; }
1223 else if(oligo[i] == 'V' && (seq[i] != 'A' && seq[i] != 'C' && seq[i] != 'G')) { countDiffs++; }
1230 catch(exception& e) {
1231 m->errorOut(e, "TrimSeqsCommand", "countDiffs");
1236 //***************************************************************************************************************
1238 //bool TrimSeqsCommand::stripQualThreshold(Sequence& seq, ifstream& qFile){
1241 // string rawSequence = seq.getUnaligned();
1242 // int seqLength = seq.getNumBases();
1243 // bool success = 0; //guilty until proven innocent
1247 // if (name[0] == '>') { if(name.substr(1) != seq.getName()) { m->mothurOut("sequence name mismatch btwn fasta: " + seq.getName() + " and qual file: " + name); m->mothurOutEndLine(); } }
1249 // while (!qFile.eof()) { char c = qFile.get(); if (c == 10 || c == 13){ break; } }
1252 // int end = seqLength;
1254 // for(int i=0;i<seqLength;i++){
1257 // if(score < qThreshold){
1262 // for(int i=end+1;i<seqLength;i++){
1266 // seq.setUnaligned(rawSequence.substr(0,end));
1270 // catch(exception& e) {
1271 // m->errorOut(e, "TrimSeqsCommand", "stripQualThreshold");
1276 //***************************************************************************************************************
1278 //bool TrimSeqsCommand::cullQualAverage(Sequence& seq, ifstream& qFile){
1280 // string rawSequence = seq.getUnaligned();
1281 // int seqLength = seq.getNumBases();
1282 // bool success = 0; //guilty until proven innocent
1286 // if (name[0] == '>') { if(name.substr(1) != seq.getName()) { m->mothurOut("sequence name mismatch btwn fasta: " + seq.getName() + " and qual file: " + name); m->mothurOutEndLine(); } }
1288 // while (!qFile.eof()) { char c = qFile.get(); if (c == 10 || c == 13){ break; } }
1291 // float average = 0;
1293 // for(int i=0;i<seqLength;i++){
1295 // average += score;
1297 // average /= seqLength;
1299 // if(average >= qAverage) { success = 1; }
1300 // else { success = 0; }
1304 // catch(exception& e) {
1305 // m->errorOut(e, "TrimSeqsCommand", "cullQualAverage");
1310 //***************************************************************************************************************