5 * Created by Pat Schloss on 12/22/10.
6 * Copyright 2010 Schloss Lab. All rights reserved.
10 #include "trimflowscommand.h"
11 #include "needlemanoverlap.hpp"
13 //**********************************************************************************************************************
15 vector<string> TrimFlowsCommand::getValidParameters(){
17 string Array[] = {"flow", "maxflows", "minflows",
18 "fasta", "minlength", "maxlength", "maxhomop", "signal", "noise"
19 "oligos", "pdiffs", "bdiffs", "tdiffs",
20 "allfiles", "processors",
21 "outputdir","inputdir"
24 vector<string> myArray (Array, Array+(sizeof(Array)/sizeof(string)));
28 m->errorOut(e, "TrimFlowsCommand", "getValidParameters");
33 //**********************************************************************************************************************
35 vector<string> TrimFlowsCommand::getRequiredParameters(){
37 string Array[] = {"flow"};
38 vector<string> myArray (Array, Array+(sizeof(Array)/sizeof(string)));
42 m->errorOut(e, "TrimFlowsCommand", "getRequiredParameters");
47 //**********************************************************************************************************************
49 vector<string> TrimFlowsCommand::getRequiredFiles(){
51 vector<string> myArray;
55 m->errorOut(e, "TrimFlowsCommand", "getRequiredFiles");
60 //**********************************************************************************************************************
62 TrimFlowsCommand::TrimFlowsCommand(){
64 abort = true; calledHelp = true;
65 vector<string> tempOutNames;
66 outputTypes["flow"] = tempOutNames;
67 outputTypes["fasta"] = tempOutNames;
70 m->errorOut(e, "TrimFlowsCommand", "TrimFlowsCommand");
75 //***************************************************************************************************************
77 TrimFlowsCommand::~TrimFlowsCommand(){ /* do nothing */ }
79 //***************************************************************************************************************
81 void TrimFlowsCommand::help(){
83 m->mothurOut("The trim.flows command reads a flowgram file and creates .....\n");
84 m->mothurOut("Note: No spaces between parameter labels (i.e. fasta), '=' and parameters (i.e.yourFasta).\n");
85 m->mothurOut("For more details please check out the wiki http://www.mothur.org/wiki/Trim.flows.\n\n");
89 m->errorOut(e, "TrimFlowsCommand", "help");
94 //**********************************************************************************************************************
96 TrimFlowsCommand::TrimFlowsCommand(string option) {
99 abort = false; calledHelp = false;
102 //allow user to run help
103 if(option == "help") { help(); abort = true; calledHelp = true; }
106 //valid paramters for this command
107 string AlignArray[] = {"flow", "maxflows", "minflows",
108 "fasta", "minlength", "maxlength", "maxhomop", "signal", "noise",
109 "oligos", "pdiffs", "bdiffs", "tdiffs",
110 "allfiles", "processors",
113 "outputdir","inputdir"
117 vector<string> myArray (AlignArray, AlignArray+(sizeof(AlignArray)/sizeof(string)));
119 OptionParser parser(option);
120 map<string,string> parameters = parser.getParameters();
122 ValidParameters validParameter;
123 map<string,string>::iterator it;
125 //check to make sure all parameters are valid for command
126 for (it = parameters.begin(); it != parameters.end(); it++) {
127 if (validParameter.isValidParameter(it->first, myArray, it->second) != true) { abort = true; }
130 //initialize outputTypes
131 vector<string> tempOutNames;
132 outputTypes["flow"] = tempOutNames;
133 outputTypes["fasta"] = tempOutNames;
135 //if the user changes the input directory command factory will send this info to us in the output parameter
136 string inputDir = validParameter.validFile(parameters, "inputdir", false);
137 if (inputDir == "not found"){ inputDir = ""; }
140 it = parameters.find("flow");
141 //user has given a template file
142 if(it != parameters.end()){
143 path = m->hasPath(it->second);
144 //if the user has not given a path then, add inputdir. else leave path alone.
145 if (path == "") { parameters["flow"] = inputDir + it->second; }
148 it = parameters.find("oligos");
149 //user has given a template file
150 if(it != parameters.end()){
151 path = m->hasPath(it->second);
152 //if the user has not given a path then, add inputdir. else leave path alone.
153 if (path == "") { parameters["oligos"] = inputDir + it->second; }
157 // it = parameters.find("group");
158 // //user has given a template file
159 // if(it != parameters.end()){
160 // path = m->hasPath(it->second);
161 // //if the user has not given a path then, add inputdir. else leave path alone.
162 // if (path == "") { parameters["group"] = inputDir + it->second; }
167 //check for required parameters
168 flowFileName = validParameter.validFile(parameters, "flow", true);
169 if (flowFileName == "not found") { m->mothurOut("flow is a required parameter for the trim.flows command."); m->mothurOutEndLine(); abort = true; }
170 else if (flowFileName == "not open") { abort = true; }
172 //if the user changes the output directory command factory will send this info to us in the output parameter
173 outputDir = validParameter.validFile(parameters, "outputdir", false); if (outputDir == "not found"){
175 outputDir += m->hasPath(flowFileName); //if user entered a file with a path then preserve it
179 //check for optional parameter and set defaults
180 // ...at some point should added some additional type checking...
183 temp = validParameter.validFile(parameters, "minflows", false); if (temp == "not found") { temp = "360"; }
184 convert(temp, minFlows);
186 temp = validParameter.validFile(parameters, "maxflows", false); if (temp == "not found") { temp = "720"; }
187 convert(temp, maxFlows);
190 temp = validParameter.validFile(parameters, "oligos", true);
191 if (temp == "not found") { oligoFileName = ""; }
192 else if(temp == "not open") { abort = true; }
193 else { oligoFileName = temp; }
195 temp = validParameter.validFile(parameters, "fasta", false); if (temp == "not found"){ fasta = 0; }
196 else if(m->isTrue(temp)) { fasta = 1; }
198 temp = validParameter.validFile(parameters, "maxhomop", false); if (temp == "not found"){ temp = "9"; }
199 convert(temp, maxHomoP);
201 temp = validParameter.validFile(parameters, "signal", false); if (temp == "not found"){ temp = "0.50"; }
202 convert(temp, signal);
204 temp = validParameter.validFile(parameters, "noise", false); if (temp == "not found"){ temp = "0.70"; }
205 convert(temp, noise);
207 temp = validParameter.validFile(parameters, "minlength", false); if (temp == "not found"){ temp = "0"; }
208 convert(temp, minLength);
210 temp = validParameter.validFile(parameters, "maxlength", false); if (temp == "not found"){ temp = "0"; }
211 convert(temp, maxLength);
213 temp = validParameter.validFile(parameters, "bdiffs", false); if (temp == "not found"){ temp = "0"; }
214 convert(temp, bdiffs);
216 temp = validParameter.validFile(parameters, "pdiffs", false); if (temp == "not found"){ temp = "0"; }
217 convert(temp, pdiffs);
219 temp = validParameter.validFile(parameters, "tdiffs", false);
220 if (temp == "not found"){ int tempTotal = pdiffs + bdiffs; temp = toString(tempTotal); }
221 convert(temp, tdiffs);
222 if(tdiffs == 0){ tdiffs = bdiffs + pdiffs; }
224 temp = validParameter.validFile(parameters, "allfiles", false); if (temp == "not found"){ temp = "T"; }
225 allFiles = m->isTrue(temp);
227 temp = validParameter.validFile(parameters, "processors", false); if (temp == "not found"){ temp = "1"; }
228 convert(temp, processors);
230 if(oligoFileName == ""){ allFiles = 0; }
237 catch(exception& e) {
238 m->errorOut(e, "TrimFlowsCommand", "TrimSeqsCommand");
243 //***************************************************************************************************************
245 int TrimFlowsCommand::execute(){
248 if (abort == true) { if (calledHelp) { return 0; } return 2; }
250 string trimFlowFileName = outputDir + m->getRootName(m->getSimpleName(flowFileName)) + "trim.flow";
251 outputNames.push_back(trimFlowFileName); outputTypes["flow"].push_back(trimFlowFileName);
253 string scrapFlowFileName = outputDir + m->getRootName(m->getSimpleName(flowFileName)) + "scrap.flow";
254 outputNames.push_back(scrapFlowFileName); outputTypes["flow"].push_back(scrapFlowFileName);
256 string fastaFileName = outputDir + m->getRootName(m->getSimpleName(flowFileName)) + "flow.fasta";
258 outputNames.push_back(fastaFileName); outputTypes["fasta"].push_back(fastaFileName);
261 vector<unsigned long int> flowFilePos = getFlowFileBreaks();
262 for (int i = 0; i < (flowFilePos.size()-1); i++) {
263 lines.push_back(new linePair(flowFilePos[i], flowFilePos[(i+1)]));
266 vector<vector<string> > barcodePrimerComboFileNames;
267 if(oligoFileName != ""){
268 getOligos(barcodePrimerComboFileNames);
271 #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
273 driverCreateTrim(flowFileName, trimFlowFileName, scrapFlowFileName, fastaFileName, barcodePrimerComboFileNames, lines[0]);
275 createProcessesCreateTrim(flowFileName, trimFlowFileName, scrapFlowFileName, fastaFileName, barcodePrimerComboFileNames);
278 driverCreateTrim(flowFileName, trimFlowFileName, scrapFlowFileName, fastaFileName, barcodePrimerComboFileNames, lines[0]);
281 if (m->control_pressed) { return 0; }
283 string flowFilesFileName;
288 flowFilesFileName = outputDir + m->getRootName(m->getSimpleName(flowFileName)) + "flow.files";
289 m->openOutputFile(flowFilesFileName, output);
291 for(int i=0;i<barcodePrimerComboFileNames.size();i++){
292 for(int j=0;j<barcodePrimerComboFileNames[0].size();j++){
295 unsigned long int size;
297 //get num bytes in file
298 pFile = fopen (barcodePrimerComboFileNames[i][j].c_str(),"rb");
299 if (pFile==NULL) perror ("Error opening file");
301 fseek (pFile, 0, SEEK_END);
307 remove(barcodePrimerComboFileNames[i][j].c_str());
310 output << barcodePrimerComboFileNames[i][j] << endl;
317 flowFilesFileName = outputDir + m->getRootName(m->getSimpleName(flowFileName)) + "flow.files";
318 m->openOutputFile(flowFilesFileName, output);
320 output << trimFlowFileName << endl;
324 outputTypes["flow.files"].push_back(flowFilesFileName);
325 outputNames.push_back(flowFileName);
327 m->mothurOutEndLine();
328 m->mothurOut("Output File Names: "); m->mothurOutEndLine();
329 for (int i = 0; i < outputNames.size(); i++) { m->mothurOut(outputNames[i]); m->mothurOutEndLine(); }
330 m->mothurOutEndLine();
334 catch(exception& e) {
335 m->errorOut(e, "TrimSeqsCommand", "execute");
340 //***************************************************************************************************************
342 int TrimFlowsCommand::driverCreateTrim(string flowFileName, string trimFlowFileName, string scrapFlowFileName, string fastaFileName, vector<vector<string> > barcodePrimerComboFileNames, linePair* line){
346 ofstream trimFlowFile;
347 m->openOutputFile(trimFlowFileName, trimFlowFile);
348 trimFlowFile.setf(ios::fixed, ios::floatfield); trimFlowFile.setf(ios::showpoint);
350 ofstream scrapFlowFile;
351 m->openOutputFile(scrapFlowFileName, scrapFlowFile);
352 scrapFlowFile.setf(ios::fixed, ios::floatfield); scrapFlowFile.setf(ios::showpoint);
354 if(line->start == 4){
355 scrapFlowFile << maxFlows << endl;
356 trimFlowFile << maxFlows << endl;
358 for(int i=0;i<barcodePrimerComboFileNames.size();i++){
359 for(int j=0;j<barcodePrimerComboFileNames[0].size();j++){
360 // barcodePrimerComboFileNames[i][j] += toString(getpid()) + ".temp";
362 m->openOutputFile(barcodePrimerComboFileNames[i][j], temp);
363 temp << maxFlows << endl;
370 FlowData flowData(numFlows, signal, noise, maxHomoP);
373 if(fasta){ m->openOutputFile(fastaFileName, fastaFile); }
376 m->openInputFile(flowFileName, flowFile);
378 flowFile.seekg(line->start);
386 int currentSeqDiffs = 0;
387 string trashCode = "";
389 flowData.getNext(flowFile);
390 flowData.capFlows(maxFlows);
392 Sequence currSeq = flowData.getSequence();
393 if(!flowData.hasMinFlows(minFlows)){ //screen to see if sequence is of a minimum number of flows
398 if(minLength > 0 || maxLength > 0){ //screen to see if sequence is above and below a specific number of bases
399 int seqLength = currSeq.getNumBases();
400 if(seqLength < minLength || seqLength > maxLength){
407 int barcodeIndex = 0;
409 if(barcodes.size() != 0){
410 success = stripBarcode(currSeq, barcodeIndex);
411 if(success > bdiffs) { trashCode += 'b'; }
412 else{ currentSeqDiffs += success; }
415 if(numFPrimers != 0){
416 success = stripForward(currSeq, primerIndex);
417 if(success > pdiffs) { trashCode += 'f'; }
418 else{ currentSeqDiffs += success; }
421 if (currentSeqDiffs > tdiffs) { trashCode += 't'; }
423 if(numRPrimers != 0){
424 success = stripReverse(currSeq);
425 if(!success) { trashCode += 'r'; }
428 if(trashCode.length() == 0){
430 flowData.printFlows(trimFlowFile);
432 if(fasta) { currSeq.printSequence(fastaFile); }
436 m->openOutputFileAppend(barcodePrimerComboFileNames[barcodeIndex][primerIndex], output);
437 output.setf(ios::fixed, ios::floatfield); trimFlowFile.setf(ios::showpoint);
439 flowData.printFlows(output);
444 flowData.printFlows(scrapFlowFile, trashCode);
450 if((count) % 10000 == 0){ m->mothurOut(toString(count)); m->mothurOutEndLine(); }
452 #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
453 unsigned long int pos = flowFile.tellg();
455 if ((pos == -1) || (pos >= line->end)) { break; }
457 if (flowFile.eof()) { break; }
462 if((count) % 10000 != 0){ m->mothurOut(toString(count)); m->mothurOutEndLine(); }
464 trimFlowFile.close();
465 scrapFlowFile.close();
467 if(fasta){ fastaFile.close(); }
471 catch(exception& e) {
472 m->errorOut(e, "TrimSeqsCommand", "driverCreateTrim");
477 //***************************************************************************************************************
479 void TrimFlowsCommand::getOligos(vector<vector<string> >& outFlowFileNames){
482 m->openInputFile(oligoFileName, oligosFile);
484 string type, oligo, group;
487 int indexBarcode = 0;
489 while(!oligosFile.eof()){
491 oligosFile >> type; m->gobble(oligosFile); //get the first column value of the row - is it a comment or a feature we are interested in?
493 if(type[0] == '#'){ //igore the line because there's a comment
494 while (!oligosFile.eof()) { char c = oligosFile.get(); if (c == 10 || c == 13){ break; } } // get rest of line if there's any crap there
496 else{ //there's a feature we're interested in
498 for(int i=0;i<type.length();i++){ type[i] = toupper(type[i]); } //make type case insensitive
500 oligosFile >> oligo; //get the DNA sequence for the feature
502 for(int i=0;i<oligo.length();i++){ //make type case insensitive and change any U's to T's
503 oligo[i] = toupper(oligo[i]);
504 if(oligo[i] == 'U') { oligo[i] = 'T'; }
507 if(type == "FORWARD"){ //if the feature is a forward primer...
510 while (!oligosFile.eof()) { // get rest of line in case there is a primer name = will have the name of the primer
511 char c = oligosFile.get();
512 if (c == 10 || c == 13){ break; }
513 else if (c == 32 || c == 9){;} //space or tab
517 //have we seen this primer already?
518 map<string, int>::iterator itPrimer = primers.find(oligo);
519 if (itPrimer != primers.end()) { m->mothurOut("primer " + oligo + " is in your oligos file already."); m->mothurOutEndLine(); }
521 primers[oligo]=indexPrimer; indexPrimer++;
522 primerNameVector.push_back(group);
525 else if(type == "REVERSE"){
526 Sequence oligoRC("reverse", oligo);
527 oligoRC.reverseComplement();
528 revPrimer.push_back(oligoRC.getUnaligned());
530 else if(type == "BARCODE"){
533 //check for repeat barcodes
534 map<string, int>::iterator itBar = barcodes.find(oligo);
535 if (itBar != barcodes.end()) { m->mothurOut("barcode " + oligo + " is in your oligos file already."); m->mothurOutEndLine(); }
537 barcodes[oligo]=indexBarcode; indexBarcode++;
538 barcodeNameVector.push_back(group);
541 m->mothurOut(type + " is not recognized as a valid type. Choices are forward, reverse, and barcode. Ignoring " + oligo + "."); m->mothurOutEndLine();
545 m->gobble(oligosFile);
549 if(barcodeNameVector.size() == 0 && primerNameVector[0] == ""){ allFiles = 0; }
551 //add in potential combos
552 if(barcodeNameVector.size() == 0){
554 barcodeNameVector.push_back("");
557 if(primerNameVector.size() == 0){
559 primerNameVector.push_back("");
563 outFlowFileNames.resize(barcodeNameVector.size());
564 for(int i=0;i<outFlowFileNames.size();i++){
565 outFlowFileNames[i].assign(primerNameVector.size(), "");
570 for(map<string, int>::iterator itBar = barcodes.begin();itBar != barcodes.end();itBar++){
571 for(map<string, int>::iterator itPrimer = primers.begin();itPrimer != primers.end(); itPrimer++){
573 string primerName = primerNameVector[itPrimer->second];
574 string barcodeName = barcodeNameVector[itBar->second];
576 string comboGroupName = "";
577 string fileName = "";
579 if(primerName == ""){
580 comboGroupName = barcodeNameVector[itBar->second];
581 fileName = outputDir + m->getRootName(m->getSimpleName(flowFileName)) + comboGroupName + ".flow";
584 if(barcodeName == ""){
585 comboGroupName = primerNameVector[itPrimer->second];
588 comboGroupName = barcodeNameVector[itBar->second] + "." + primerNameVector[itPrimer->second];
590 fileName = outputDir + m->getRootName(m->getSimpleName(flowFileName)) + comboGroupName + ".flow";
593 outputNames.push_back(fileName);
594 outputTypes["flow"].push_back(fileName);
595 outFlowFileNames[itBar->second][itPrimer->second] = fileName;
598 m->openOutputFile(fileName, temp);
604 numFPrimers = primers.size();
605 numRPrimers = revPrimer.size();
608 catch(exception& e) {
609 m->errorOut(e, "TrimSeqsCommand", "getOligos");
614 //***************************************************************************************************************
616 int TrimFlowsCommand::stripBarcode(Sequence& seq, int& group){
619 string rawSequence = seq.getUnaligned();
620 int success = bdiffs + 1; //guilty until proven innocent
622 //can you find the barcode
623 for(map<string,int>::iterator it=barcodes.begin();it!=barcodes.end();it++){
624 string oligo = it->first;
625 if(rawSequence.length() < oligo.length()){ //let's just assume that the barcodes are the same length
626 success = bdiffs + 10; //if the sequence is shorter than the barcode then bail out
630 if(compareDNASeq(oligo, rawSequence.substr(0,oligo.length()))){
632 seq.setUnaligned(rawSequence.substr(oligo.length()));
639 //if you found the barcode or if you don't want to allow for diffs
640 if ((bdiffs == 0) || (success == 0)) { return success; }
642 else { //try aligning and see if you can find it
646 Alignment* alignment;
647 if (barcodes.size() > 0) {
648 map<string,int>::iterator it=barcodes.begin();
650 for(it;it!=barcodes.end();it++){
651 if(it->first.length() > maxLength){
652 maxLength = it->first.length();
655 alignment = new NeedlemanOverlap(-1.0, 1.0, -1.0, (maxLength+bdiffs+1));
657 }else{ alignment = NULL; }
659 //can you find the barcode
665 for(map<string,int>::iterator it=barcodes.begin();it!=barcodes.end();it++){
666 string oligo = it->first;
667 // int length = oligo.length();
669 if(rawSequence.length() < maxLength){ //let's just assume that the barcodes are the same length
670 success = bdiffs + 10;
674 //use needleman to align first barcode.length()+numdiffs of sequence to each barcode
675 alignment->align(oligo, rawSequence.substr(0,oligo.length()+bdiffs));
676 oligo = alignment->getSeqAAln();
677 string temp = alignment->getSeqBAln();
679 int alnLength = oligo.length();
681 for(int i=oligo.length()-1;i>=0;i--){
682 if(oligo[i] != '-'){ alnLength = i+1; break; }
684 oligo = oligo.substr(0,alnLength);
685 temp = temp.substr(0,alnLength);
687 int numDiff = countDiffs(oligo, temp);
689 if(numDiff < minDiff){
692 minGroup = it->second;
694 for(int i=0;i<alnLength;i++){
700 else if(numDiff == minDiff){
706 if(minDiff > bdiffs) { success = minDiff; } //no good matches
707 else if(minCount > 1) { success = bdiffs + 100; } //can't tell the difference between multiple barcodes
708 else{ //use the best match
710 seq.setUnaligned(rawSequence.substr(minPos));
714 if (alignment != NULL) { delete alignment; }
721 catch(exception& e) {
722 m->errorOut(e, "TrimFlowsCommand", "stripBarcode");
728 //***************************************************************************************************************
730 int TrimFlowsCommand::stripForward(Sequence& seq, int& group){
733 string rawSequence = seq.getUnaligned();
734 int success = pdiffs + 1; //guilty until proven innocent
736 //can you find the primer
737 for(map<string,int>::iterator it=primers.begin();it!=primers.end();it++){
738 string oligo = it->first;
739 if(rawSequence.length() < oligo.length()){ //let's just assume that the primers are the same length
740 success = pdiffs + 10; //if the sequence is shorter than the barcode then bail out
744 if(compareDNASeq(oligo, rawSequence.substr(0,oligo.length()))){
746 seq.setUnaligned(rawSequence.substr(oligo.length()));
752 //if you found the barcode or if you don't want to allow for diffs
753 if ((pdiffs == 0) || (success == 0)) { return success; }
755 else { //try aligning and see if you can find it
759 Alignment* alignment;
760 if (primers.size() > 0) {
761 map<string,int>::iterator it=primers.begin();
763 for(it;it!=primers.end();it++){
764 if(it->first.length() > maxLength){
765 maxLength = it->first.length();
768 alignment = new NeedlemanOverlap(-1.0, 1.0, -1.0, (maxLength+pdiffs+1));
770 }else{ alignment = NULL; }
772 //can you find the barcode
778 for(map<string,int>::iterator it=primers.begin();it!=primers.end();it++){
779 string oligo = it->first;
780 // int length = oligo.length();
782 if(rawSequence.length() < maxLength){
783 success = pdiffs + 100;
787 //use needleman to align first barcode.length()+numdiffs of sequence to each barcode
788 alignment->align(oligo, rawSequence.substr(0,oligo.length()+pdiffs));
789 oligo = alignment->getSeqAAln();
790 string temp = alignment->getSeqBAln();
792 int alnLength = oligo.length();
794 for(int i=oligo.length()-1;i>=0;i--){
795 if(oligo[i] != '-'){ alnLength = i+1; break; }
797 oligo = oligo.substr(0,alnLength);
798 temp = temp.substr(0,alnLength);
800 int numDiff = countDiffs(oligo, temp);
802 if(numDiff < minDiff){
805 minGroup = it->second;
807 for(int i=0;i<alnLength;i++){
813 else if(numDiff == minDiff){
819 if(minDiff > pdiffs) { success = minDiff; } //no good matches
820 else if(minCount > 1) { success = pdiffs + 10; } //can't tell the difference between multiple primers
821 else{ //use the best match
823 seq.setUnaligned(rawSequence.substr(minPos));
827 if (alignment != NULL) { delete alignment; }
834 catch(exception& e) {
835 m->errorOut(e, "TrimFlowsCommand", "stripForward");
840 //***************************************************************************************************************
842 bool TrimFlowsCommand::stripReverse(Sequence& seq){
845 string rawSequence = seq.getUnaligned();
846 bool success = 0; //guilty until proven innocent
848 for(int i=0;i<numRPrimers;i++){
849 string oligo = revPrimer[i];
851 if(rawSequence.length() < oligo.length()){
856 if(compareDNASeq(oligo, rawSequence.substr(rawSequence.length()-oligo.length(),oligo.length()))){
857 seq.setUnaligned(rawSequence.substr(0,rawSequence.length()-oligo.length()));
866 catch(exception& e) {
867 m->errorOut(e, "TrimFlowsCommand", "stripReverse");
873 //***************************************************************************************************************
875 bool TrimFlowsCommand::compareDNASeq(string oligo, string seq){
878 int length = oligo.length();
880 for(int i=0;i<length;i++){
882 if(oligo[i] != seq[i]){
883 if(oligo[i] == 'A' || oligo[i] == 'T' || oligo[i] == 'G' || oligo[i] == 'C') { success = 0; }
884 else if((oligo[i] == 'N' || oligo[i] == 'I') && (seq[i] == 'N')) { success = 0; }
885 else if(oligo[i] == 'R' && (seq[i] != 'A' && seq[i] != 'G')) { success = 0; }
886 else if(oligo[i] == 'Y' && (seq[i] != 'C' && seq[i] != 'T')) { success = 0; }
887 else if(oligo[i] == 'M' && (seq[i] != 'C' && seq[i] != 'A')) { success = 0; }
888 else if(oligo[i] == 'K' && (seq[i] != 'T' && seq[i] != 'G')) { success = 0; }
889 else if(oligo[i] == 'W' && (seq[i] != 'T' && seq[i] != 'A')) { success = 0; }
890 else if(oligo[i] == 'S' && (seq[i] != 'C' && seq[i] != 'G')) { success = 0; }
891 else if(oligo[i] == 'B' && (seq[i] != 'C' && seq[i] != 'T' && seq[i] != 'G')) { success = 0; }
892 else if(oligo[i] == 'D' && (seq[i] != 'A' && seq[i] != 'T' && seq[i] != 'G')) { success = 0; }
893 else if(oligo[i] == 'H' && (seq[i] != 'A' && seq[i] != 'T' && seq[i] != 'C')) { success = 0; }
894 else if(oligo[i] == 'V' && (seq[i] != 'A' && seq[i] != 'C' && seq[i] != 'G')) { success = 0; }
896 if(success == 0) { break; }
905 catch(exception& e) {
906 m->errorOut(e, "TrimFlowsCommand", "compareDNASeq");
912 //***************************************************************************************************************
914 int TrimFlowsCommand::countDiffs(string oligo, string seq){
917 int length = oligo.length();
920 for(int i=0;i<length;i++){
922 if(oligo[i] != seq[i]){
923 if(oligo[i] == 'A' || oligo[i] == 'T' || oligo[i] == 'G' || oligo[i] == 'C' || oligo[i] == '-' || oligo[i] == '.') { countDiffs++; }
924 else if((oligo[i] == 'N' || oligo[i] == 'I') && (seq[i] == 'N')) { countDiffs++; }
925 else if(oligo[i] == 'R' && (seq[i] != 'A' && seq[i] != 'G')) { countDiffs++; }
926 else if(oligo[i] == 'Y' && (seq[i] != 'C' && seq[i] != 'T')) { countDiffs++; }
927 else if(oligo[i] == 'M' && (seq[i] != 'C' && seq[i] != 'A')) { countDiffs++; }
928 else if(oligo[i] == 'K' && (seq[i] != 'T' && seq[i] != 'G')) { countDiffs++; }
929 else if(oligo[i] == 'W' && (seq[i] != 'T' && seq[i] != 'A')) { countDiffs++; }
930 else if(oligo[i] == 'S' && (seq[i] != 'C' && seq[i] != 'G')) { countDiffs++; }
931 else if(oligo[i] == 'B' && (seq[i] != 'C' && seq[i] != 'T' && seq[i] != 'G')) { countDiffs++; }
932 else if(oligo[i] == 'D' && (seq[i] != 'A' && seq[i] != 'T' && seq[i] != 'G')) { countDiffs++; }
933 else if(oligo[i] == 'H' && (seq[i] != 'A' && seq[i] != 'T' && seq[i] != 'C')) { countDiffs++; }
934 else if(oligo[i] == 'V' && (seq[i] != 'A' && seq[i] != 'C' && seq[i] != 'G')) { countDiffs++; }
941 catch(exception& e) {
942 m->errorOut(e, "TrimFlowsCommand", "countDiffs");
948 /**************************************************************************************************/
950 vector<unsigned long int> TrimFlowsCommand::getFlowFileBreaks() {
954 vector<unsigned long int> filePos;
955 filePos.push_back(0);
958 unsigned long int size;
960 //get num bytes in file
961 pFile = fopen (flowFileName.c_str(),"rb");
962 if (pFile==NULL) perror ("Error opening file");
964 fseek (pFile, 0, SEEK_END);
969 //estimate file breaks
970 unsigned long int chunkSize = 0;
971 chunkSize = size / processors;
973 //file too small to divide by processors
974 if (chunkSize == 0) { processors = 1; filePos.push_back(size); return filePos; }
976 //for each process seekg to closest file break and search for next '>' char. make that the filebreak
977 for (int i = 0; i < processors; i++) {
978 unsigned long int spot = (i+1) * chunkSize;
981 m->openInputFile(flowFileName, in);
984 string dummy = m->getline(in);
986 //there was not another sequence before the end of the file
987 unsigned long int sanityPos = in.tellg();
989 // if (sanityPos == -1) { break; }
990 // else { filePos.push_back(newSpot); }
991 if (sanityPos == -1) { break; }
992 else { filePos.push_back(sanityPos); }
998 filePos.push_back(size);
1000 //sanity check filePos
1001 for (int i = 0; i < (filePos.size()-1); i++) {
1002 if (filePos[(i+1)] <= filePos[i]) { filePos.erase(filePos.begin()+(i+1)); i--; }
1006 m->openInputFile(flowFileName, in);
1009 unsigned long int spot = in.tellg();
1013 processors = (filePos.size() - 1);
1017 catch(exception& e) {
1018 m->errorOut(e, "TrimSeqsCommand", "getFlowFileBreaks");
1023 /**************************************************************************************************/
1025 int TrimFlowsCommand::createProcessesCreateTrim(string flowFileName, string trimFlowFileName, string scrapFlowFileName, string fastaFileName, vector<vector<string> > barcodePrimerComboFileNames){
1028 #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
1030 int exitCommand = 1;
1033 //loop through and create all the processes you want
1034 while (process != processors) {
1038 processIDS.push_back(pid); //create map from line number to pid so you can append files in correct order later
1040 }else if (pid == 0){
1042 vector<vector<string> > tempBarcodePrimerComboFileNames = barcodePrimerComboFileNames;
1044 for(int i=0;i<tempBarcodePrimerComboFileNames.size();i++){
1045 for(int j=0;j<tempBarcodePrimerComboFileNames[0].size();j++){
1046 tempBarcodePrimerComboFileNames[i][j] += toString(getpid()) + ".temp";
1048 m->openOutputFile(tempBarcodePrimerComboFileNames[i][j], temp);
1054 driverCreateTrim(flowFileName,
1055 (trimFlowFileName + toString(getpid()) + ".temp"),
1056 (scrapFlowFileName + toString(getpid()) + ".temp"),
1057 (fastaFileName + toString(getpid()) + ".temp"),
1058 tempBarcodePrimerComboFileNames, lines[process]);
1062 m->mothurOut("[ERROR]: unable to spawn the necessary processes."); m->mothurOutEndLine();
1063 for (int i = 0; i < processIDS.size(); i++) { kill (processIDS[i], SIGINT); }
1070 m->openOutputFile(trimFlowFileName, temp);
1073 m->openOutputFile(scrapFlowFileName, temp);
1077 m->openOutputFile(fastaFileName, temp);
1081 driverCreateTrim(flowFileName, trimFlowFileName, scrapFlowFileName, fastaFileName, barcodePrimerComboFileNames, lines[0]);
1083 //force parent to wait until all the processes are done
1084 for (int i=0;i<processIDS.size();i++) {
1085 int temp = processIDS[i];
1090 m->mothurOutEndLine();
1091 for(int i=0;i<processIDS.size();i++){
1093 m->mothurOut("Appending files from process " + toString(processIDS[i])); m->mothurOutEndLine();
1095 m->appendFiles((trimFlowFileName + toString(processIDS[i]) + ".temp"), trimFlowFileName);
1096 remove((trimFlowFileName + toString(processIDS[i]) + ".temp").c_str());
1097 // m->mothurOut("\tDone with trim.flow file"); m->mothurOutEndLine();
1099 m->appendFiles((scrapFlowFileName + toString(processIDS[i]) + ".temp"), scrapFlowFileName);
1100 remove((scrapFlowFileName + toString(processIDS[i]) + ".temp").c_str());
1101 // m->mothurOut("\tDone with scrap.flow file"); m->mothurOutEndLine();
1104 m->appendFiles((fastaFileName + toString(processIDS[i]) + ".temp"), fastaFileName);
1105 remove((fastaFileName + toString(processIDS[i]) + ".temp").c_str());
1106 // m->mothurOut("\tDone with flow.fasta file"); m->mothurOutEndLine();
1109 for (int j = 0; j < barcodePrimerComboFileNames.size(); j++) {
1110 for (int k = 0; k < barcodePrimerComboFileNames[0].size(); k++) {
1111 m->appendFiles((barcodePrimerComboFileNames[j][k] + toString(processIDS[i]) + ".temp"), barcodePrimerComboFileNames[j][k]);
1112 remove((barcodePrimerComboFileNames[j][k] + toString(processIDS[i]) + ".temp").c_str());
1121 catch(exception& e) {
1122 m->errorOut(e, "TrimFlowsCommand", "createProcessesCreateTrim");
1127 //***************************************************************************************************************