2 * removelineagecommand.cpp
5 * Created by westcott on 9/24/10.
6 * Copyright 2010 Schloss Lab. All rights reserved.
10 #include "removelineagecommand.h"
11 #include "sequence.hpp"
12 #include "listvector.hpp"
14 //**********************************************************************************************************************
15 vector<string> RemoveLineageCommand::getValidParameters(){
17 string Array[] = {"fasta","name", "group", "alignreport", "taxon", "dups", "list","taxonomy","outputdir","inputdir"};
18 vector<string> myArray (Array, Array+(sizeof(Array)/sizeof(string)));
22 m->errorOut(e, "RemoveLineageCommand", "getValidParameters");
26 //**********************************************************************************************************************
27 RemoveLineageCommand::RemoveLineageCommand(){
29 abort = true; calledHelp = true;
30 vector<string> tempOutNames;
31 outputTypes["fasta"] = tempOutNames;
32 outputTypes["taxonomy"] = tempOutNames;
33 outputTypes["name"] = tempOutNames;
34 outputTypes["group"] = tempOutNames;
35 outputTypes["alignreport"] = tempOutNames;
36 outputTypes["list"] = tempOutNames;
39 m->errorOut(e, "RemoveLineageCommand", "RemoveLineageCommand");
43 //**********************************************************************************************************************
44 vector<string> RemoveLineageCommand::getRequiredParameters(){
46 string Array[] = {"taxonomy"};
47 vector<string> myArray (Array, Array+(sizeof(Array)/sizeof(string)));
51 m->errorOut(e, "RemoveLineageCommand", "getRequiredParameters");
55 //**********************************************************************************************************************
56 vector<string> RemoveLineageCommand::getRequiredFiles(){
58 vector<string> myArray;
62 m->errorOut(e, "RemoveLineageCommand", "getRequiredFiles");
66 //**********************************************************************************************************************
67 RemoveLineageCommand::RemoveLineageCommand(string option) {
69 abort = false; calledHelp = false;
71 //allow user to run help
72 if(option == "help") { help(); abort = true; calledHelp = true; }
75 //valid paramters for this command
76 string Array[] = {"fasta","name", "group", "alignreport", "taxon", "dups", "list","taxonomy","outputdir","inputdir"};
77 vector<string> myArray (Array, Array+(sizeof(Array)/sizeof(string)));
79 OptionParser parser(option);
80 map<string,string> parameters = parser.getParameters();
82 ValidParameters validParameter;
83 map<string,string>::iterator it;
85 //check to make sure all parameters are valid for command
86 for (it = parameters.begin(); it != parameters.end(); it++) {
87 if (validParameter.isValidParameter(it->first, myArray, it->second) != true) { abort = true; }
90 //initialize outputTypes
91 vector<string> tempOutNames;
92 outputTypes["fasta"] = tempOutNames;
93 outputTypes["taxonomy"] = tempOutNames;
94 outputTypes["name"] = tempOutNames;
95 outputTypes["group"] = tempOutNames;
96 outputTypes["alignreport"] = tempOutNames;
97 outputTypes["list"] = tempOutNames;
99 //if the user changes the output directory command factory will send this info to us in the output parameter
100 outputDir = validParameter.validFile(parameters, "outputdir", false); if (outputDir == "not found"){ outputDir = ""; }
102 //if the user changes the input directory command factory will send this info to us in the output parameter
103 string inputDir = validParameter.validFile(parameters, "inputdir", false);
104 if (inputDir == "not found"){ inputDir = ""; }
107 it = parameters.find("alignreport");
108 //user has given a template file
109 if(it != parameters.end()){
110 path = m->hasPath(it->second);
111 //if the user has not given a path then, add inputdir. else leave path alone.
112 if (path == "") { parameters["alignreport"] = inputDir + it->second; }
115 it = parameters.find("fasta");
116 //user has given a template file
117 if(it != parameters.end()){
118 path = m->hasPath(it->second);
119 //if the user has not given a path then, add inputdir. else leave path alone.
120 if (path == "") { parameters["fasta"] = inputDir + it->second; }
123 it = parameters.find("list");
124 //user has given a template file
125 if(it != parameters.end()){
126 path = m->hasPath(it->second);
127 //if the user has not given a path then, add inputdir. else leave path alone.
128 if (path == "") { parameters["list"] = inputDir + it->second; }
131 it = parameters.find("name");
132 //user has given a template file
133 if(it != parameters.end()){
134 path = m->hasPath(it->second);
135 //if the user has not given a path then, add inputdir. else leave path alone.
136 if (path == "") { parameters["name"] = inputDir + it->second; }
139 it = parameters.find("group");
140 //user has given a template file
141 if(it != parameters.end()){
142 path = m->hasPath(it->second);
143 //if the user has not given a path then, add inputdir. else leave path alone.
144 if (path == "") { parameters["group"] = inputDir + it->second; }
147 it = parameters.find("taxonomy");
148 //user has given a template file
149 if(it != parameters.end()){
150 path = m->hasPath(it->second);
151 //if the user has not given a path then, add inputdir. else leave path alone.
152 if (path == "") { parameters["taxonomy"] = inputDir + it->second; }
157 //check for required parameters
158 fastafile = validParameter.validFile(parameters, "fasta", true);
159 if (fastafile == "not open") { abort = true; }
160 else if (fastafile == "not found") { fastafile = ""; }
162 namefile = validParameter.validFile(parameters, "name", true);
163 if (namefile == "not open") { abort = true; }
164 else if (namefile == "not found") { namefile = ""; }
166 groupfile = validParameter.validFile(parameters, "group", true);
167 if (groupfile == "not open") { abort = true; }
168 else if (groupfile == "not found") { groupfile = ""; }
170 alignfile = validParameter.validFile(parameters, "alignreport", true);
171 if (alignfile == "not open") { abort = true; }
172 else if (alignfile == "not found") { alignfile = ""; }
174 listfile = validParameter.validFile(parameters, "list", true);
175 if (listfile == "not open") { abort = true; }
176 else if (listfile == "not found") { listfile = ""; }
178 taxfile = validParameter.validFile(parameters, "taxonomy", true);
179 if (taxfile == "not open") { abort = true; }
180 else if (taxfile == "not found") { taxfile = ""; m->mothurOut("The taxonomy parameter is required for the get.lineage command."); m->mothurOutEndLine(); abort = true; }
182 string usedDups = "true";
183 string temp = validParameter.validFile(parameters, "dups", false);
184 if (temp == "not found") {
185 if (namefile != "") { temp = "true"; }
186 else { temp = "false"; usedDups = ""; }
188 dups = m->isTrue(temp);
190 taxons = validParameter.validFile(parameters, "taxon", false);
191 if (taxons == "not found") { taxons = ""; m->mothurOut("No taxons given, please correct."); m->mothurOutEndLine(); abort = true; }
194 if (taxons[0] == '\'') { taxons = taxons.substr(1); }
195 if (taxons[(taxons.length()-1)] == '\'') { taxons = taxons.substr(0, (taxons.length()-1)); }
199 if ((fastafile == "") && (namefile == "") && (groupfile == "") && (alignfile == "") && (listfile == "") && (taxfile == "")) { m->mothurOut("You must provide one of the following: fasta, name, group, alignreport, taxonomy or listfile."); m->mothurOutEndLine(); abort = true; }
201 if ((usedDups != "") && (namefile == "")) { m->mothurOut("You may only use dups with the name option."); m->mothurOutEndLine(); abort = true; }
206 catch(exception& e) {
207 m->errorOut(e, "RemoveLineageCommand", "RemoveLineageCommand");
211 //**********************************************************************************************************************
213 void RemoveLineageCommand::help(){
215 m->mothurOut("The remove.lineage command reads a taxonomy file and any of the following file types: fasta, name, group, list or alignreport file.\n");
216 m->mothurOut("It outputs a file containing only the sequences from the taxonomy file that are not from the taxon you requested to be removed.\n");
217 m->mothurOut("The remove.lineage command parameters are taxon, fasta, name, group, list, taxonomy, alignreport and dups. You must provide taxonomy and taxon.\n");
218 m->mothurOut("The dups parameter allows you to add the entire line from a name file if you add any name from the line. default=false. \n");
219 m->mothurOut("The taxon parameter allows you to select the taxons you would like to remove.\n");
220 m->mothurOut("You may enter your taxons with confidence scores, doing so will remove only those sequences that belong to the taxonomy and whose cofidence scores fall below the scores you give.\n");
221 m->mothurOut("If they belong to the taxonomy and have confidences above those you provide the sequence will not be removed.\n");
222 m->mothurOut("The remove.lineage command should be in the following format: remove.lineage(taxonomy=yourTaxonomyFile, taxon=yourTaxons).\n");
223 m->mothurOut("Example remove.lineage(taxonomy=amazon.silva.taxonomy, taxon=Bacteria;Firmicutes;Bacilli;Lactobacillales;).\n");
224 m->mothurOut("Note: If you are running mothur in script mode you must wrap the taxon in ' characters so mothur will ignore the ; in the taxon.\n");
225 m->mothurOut("Example remove.lineage(taxonomy=amazon.silva.taxonomy, taxon='Bacteria;Firmicutes;Bacilli;Lactobacillales;').\n");
226 m->mothurOut("Note: No spaces between parameter labels (i.e. fasta), '=' and parameters (i.e.yourFasta).\n\n");
228 catch(exception& e) {
229 m->errorOut(e, "RemoveLineageCommand", "help");
234 //**********************************************************************************************************************
236 int RemoveLineageCommand::execute(){
239 if (abort == true) { if (calledHelp) { return 0; } return 2; }
241 if (m->control_pressed) { return 0; }
243 //read through the correct file and output lines you want to keep
244 if (taxfile != "") { readTax(); } //fills the set of names to remove
245 if (namefile != "") { readName(); }
246 if (fastafile != "") { readFasta(); }
247 if (groupfile != "") { readGroup(); }
248 if (alignfile != "") { readAlign(); }
249 if (listfile != "") { readList(); }
252 if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } return 0; }
254 if (outputNames.size() != 0) {
255 m->mothurOutEndLine();
256 m->mothurOut("Output File Names: "); m->mothurOutEndLine();
257 for (int i = 0; i < outputNames.size(); i++) { m->mothurOut(outputNames[i]); m->mothurOutEndLine(); }
258 m->mothurOutEndLine();
264 catch(exception& e) {
265 m->errorOut(e, "RemoveLineageCommand", "execute");
270 //**********************************************************************************************************************
271 int RemoveLineageCommand::readFasta(){
273 string thisOutputDir = outputDir;
274 if (outputDir == "") { thisOutputDir += m->hasPath(fastafile); }
275 string outputFileName = thisOutputDir + m->getRootName(m->getSimpleName(fastafile)) + "pick" + m->getExtension(fastafile);
278 m->openOutputFile(outputFileName, out);
281 m->openInputFile(fastafile, in);
284 bool wroteSomething = false;
287 if (m->control_pressed) { in.close(); out.close(); remove(outputFileName.c_str()); return 0; }
289 Sequence currSeq(in);
290 name = currSeq.getName();
293 //if this name is in the accnos file
294 if (names.count(name) == 0) {
295 wroteSomething = true;
297 currSeq.printSequence(out);
305 if (wroteSomething == false) { m->mothurOut("Your fasta file contains only sequences from " + taxons + "."); m->mothurOutEndLine(); }
306 outputNames.push_back(outputFileName); outputTypes["fasta"].push_back(outputFileName);
311 catch(exception& e) {
312 m->errorOut(e, "RemoveLineageCommand", "readFasta");
316 //**********************************************************************************************************************
317 int RemoveLineageCommand::readList(){
319 string thisOutputDir = outputDir;
320 if (outputDir == "") { thisOutputDir += m->hasPath(listfile); }
321 string outputFileName = thisOutputDir + m->getRootName(m->getSimpleName(listfile)) + "pick" + m->getExtension(listfile);
324 m->openOutputFile(outputFileName, out);
327 m->openInputFile(listfile, in);
329 bool wroteSomething = false;
332 //read in list vector
335 //make a new list vector
337 newList.setLabel(list.getLabel());
340 for (int i = 0; i < list.getNumBins(); i++) {
341 if (m->control_pressed) { in.close(); out.close(); remove(outputFileName.c_str()); return 0; }
343 //parse out names that are in accnos file
344 string binnames = list.get(i);
346 string newNames = "";
347 while (binnames.find_first_of(',') != -1) {
348 string name = binnames.substr(0,binnames.find_first_of(','));
349 binnames = binnames.substr(binnames.find_first_of(',')+1, binnames.length());
351 //if that name is in the .accnos file, add it
352 if (names.count(name) == 0) { newNames += name + ","; }
356 if (names.count(binnames) == 0) { newNames += binnames + ","; }
358 //if there are names in this bin add to new list
359 if (newNames != "") {
360 newNames = newNames.substr(0, newNames.length()-1); //rip off extra comma
361 newList.push_back(newNames);
365 //print new listvector
366 if (newList.getNumBins() != 0) {
367 wroteSomething = true;
376 if (wroteSomething == false) { m->mothurOut("Your list file contains only sequences from " + taxons + "."); m->mothurOutEndLine(); }
377 outputNames.push_back(outputFileName); outputTypes["list"].push_back(outputFileName);
382 catch(exception& e) {
383 m->errorOut(e, "RemoveLineageCommand", "readList");
387 //**********************************************************************************************************************
388 int RemoveLineageCommand::readName(){
390 string thisOutputDir = outputDir;
391 if (outputDir == "") { thisOutputDir += m->hasPath(namefile); }
392 string outputFileName = thisOutputDir + m->getRootName(m->getSimpleName(namefile)) + "pick" + m->getExtension(namefile);
395 m->openOutputFile(outputFileName, out);
398 m->openInputFile(namefile, in);
399 string name, firstCol, secondCol;
401 bool wroteSomething = false;
404 if (m->control_pressed) { in.close(); out.close(); remove(outputFileName.c_str()); return 0; }
409 vector<string> parsedNames;
410 //parse second column saving each name
411 while (secondCol.find_first_of(',') != -1) {
412 name = secondCol.substr(0,secondCol.find_first_of(','));
413 secondCol = secondCol.substr(secondCol.find_first_of(',')+1, secondCol.length());
414 parsedNames.push_back(name);
417 //get name after last ,
418 parsedNames.push_back(secondCol);
420 vector<string> validSecond; validSecond.clear();
421 for (int i = 0; i < parsedNames.size(); i++) {
422 if (names.count(parsedNames[i]) == 0) {
423 validSecond.push_back(parsedNames[i]);
427 if ((dups) && (validSecond.size() != parsedNames.size())) { //if dups is true and we want to get rid of anyone, get rid of everyone
428 for (int i = 0; i < parsedNames.size(); i++) { names.insert(parsedNames[i]); }
430 //if the name in the first column is in the set then print it and any other names in second column also in set
431 if (names.count(firstCol) == 0) {
433 wroteSomething = true;
435 out << firstCol << '\t';
437 //you know you have at least one valid second since first column is valid
438 for (int i = 0; i < validSecond.size()-1; i++) { out << validSecond[i] << ','; }
439 out << validSecond[validSecond.size()-1] << endl;
441 //make first name in set you come to first column and then add the remaining names to second column
444 //you want part of this row
445 if (validSecond.size() != 0) {
447 wroteSomething = true;
449 out << validSecond[0] << '\t';
451 //you know you have at least one valid second since first column is valid
452 for (int i = 0; i < validSecond.size()-1; i++) { out << validSecond[i] << ','; }
453 out << validSecond[validSecond.size()-1] << endl;
462 if (wroteSomething == false) { m->mothurOut("Your name file contains only sequences from " + taxons + "."); m->mothurOutEndLine(); }
463 outputNames.push_back(outputFileName); outputTypes["name"].push_back(outputFileName);
467 catch(exception& e) {
468 m->errorOut(e, "RemoveLineageCommand", "readName");
473 //**********************************************************************************************************************
474 int RemoveLineageCommand::readGroup(){
476 string thisOutputDir = outputDir;
477 if (outputDir == "") { thisOutputDir += m->hasPath(groupfile); }
478 string outputFileName = thisOutputDir + m->getRootName(m->getSimpleName(groupfile)) + "pick" + m->getExtension(groupfile);
481 m->openOutputFile(outputFileName, out);
484 m->openInputFile(groupfile, in);
487 bool wroteSomething = false;
490 if (m->control_pressed) { in.close(); out.close(); remove(outputFileName.c_str()); return 0; }
492 in >> name; //read from first column
493 in >> group; //read from second column
495 //if this name is in the accnos file
496 if (names.count(name) == 0) {
497 wroteSomething = true;
498 out << name << '\t' << group << endl;
506 if (wroteSomething == false) { m->mothurOut("Your group file contains only sequences from " + taxons + "."); m->mothurOutEndLine(); }
507 outputNames.push_back(outputFileName); outputTypes["group"].push_back(outputFileName);
511 catch(exception& e) {
512 m->errorOut(e, "RemoveLineageCommand", "readGroup");
516 //**********************************************************************************************************************
517 int RemoveLineageCommand::readTax(){
519 string thisOutputDir = outputDir;
520 if (outputDir == "") { thisOutputDir += m->hasPath(taxfile); }
521 string outputFileName = thisOutputDir + m->getRootName(m->getSimpleName(taxfile)) + "pick" + m->getExtension(taxfile);
523 m->openOutputFile(outputFileName, out);
526 m->openInputFile(taxfile, in);
529 bool wroteSomething = false;
531 bool taxonsHasConfidence = false;
532 vector< map<string, int> > searchTaxons;
533 string noConfidenceTaxons = taxons;
534 int hasConPos = taxons.find_first_of('(');
535 if (hasConPos != string::npos) {
536 taxonsHasConfidence = true;
537 searchTaxons = getTaxons(taxons);
538 noConfidenceTaxons = removeConfidences(taxons);
544 if (m->control_pressed) { in.close(); out.close(); remove(outputFileName.c_str()); return 0; }
546 in >> name; //read from first column
547 in >> tax; //read from second column
551 //if the users file contains confidence scores we want to ignore them when searching for the taxons, unless the taxon has them
552 if (!taxonsHasConfidence) {
554 int hasConfidences = tax.find_first_of('(');
555 if (hasConfidences != string::npos) {
556 newtax = removeConfidences(tax);
559 int pos = newtax.find(taxons);
561 if (pos == string::npos) {
562 wroteSomething = true;
563 out << name << '\t' << tax << endl;
564 }else{ //this sequence contains the taxon the user wants to remove
568 }else{//if taxons has them and you don't them remove taxons
569 int hasConfidences = tax.find_first_of('(');
570 if (hasConfidences == string::npos) {
572 int pos = newtax.find(noConfidenceTaxons);
574 if (pos == string::npos) {
575 wroteSomething = true;
576 out << name << '\t' << tax << endl;
577 }else{ //this sequence contains the taxon the user wants to remove
580 }else { //both have confidences so we want to make sure the users confidences are greater then or equal to the taxons
581 //first remove confidences from both and see if the taxonomy exists
583 string noNewTax = tax;
584 int hasConfidences = tax.find_first_of('(');
585 if (hasConfidences != string::npos) {
586 noNewTax = removeConfidences(tax);
589 int pos = noNewTax.find(noConfidenceTaxons);
591 if (pos != string::npos) { //if yes, then are the confidences okay
594 vector< map<string, int> > usersTaxon = getTaxons(newtax);
596 //the usersTaxon is most likely longer than the searchTaxons, and searchTaxon[0] may relate to userTaxon[4]
597 //we want to "line them up", so we will find the the index where the searchstring starts
599 for (int i = 0; i < usersTaxon.size(); i++) {
601 if (usersTaxon[i].begin()->first == searchTaxons[0].begin()->first) {
604 bool goodspot = true;
605 //is this really the start, or are we dealing with a taxon of the same name?
606 while ((spot < searchTaxons.size()) && ((i+spot) < usersTaxon.size())) {
607 if (usersTaxon[i+spot].begin()->first != searchTaxons[spot].begin()->first) { goodspot = false; break; }
611 if (goodspot) { break; }
615 for (int i = 0; i < searchTaxons.size(); i++) {
617 if ((i+index) < usersTaxon.size()) { //just in case, should never be false
618 if (usersTaxon[i+index].begin()->second < searchTaxons[i].begin()->second) { //is the users cutoff less than the search taxons
628 //passed the test so remove you
632 wroteSomething = true;
633 out << name << '\t' << tax << endl;
636 wroteSomething = true;
637 out << name << '\t' << tax << endl;
650 if (!wroteSomething) { m->mothurOut("Your taxonomy file contains only sequences from " + taxons + "."); m->mothurOutEndLine(); }
651 outputNames.push_back(outputFileName); outputTypes["taxonomy"].push_back(outputFileName);
656 catch(exception& e) {
657 m->errorOut(e, "RemoveLineageCommand", "readTax");
661 /**************************************************************************************************/
662 vector< map<string, int> > RemoveLineageCommand::getTaxons(string tax) {
665 vector< map<string, int> > t;
667 int taxLength = tax.length();
668 for(int i=0;i<taxLength;i++){
671 int openParen = taxon.find_first_of('(');
672 int closeParen = taxon.find_last_of(')');
674 string newtaxon, confidence;
675 if ((openParen != string::npos) && (closeParen != string::npos)) {
676 newtaxon = taxon.substr(0, openParen); //rip off confidence
677 confidence = taxon.substr((openParen+1), (closeParen-openParen-1));
683 convert(confidence, con);
685 map<string, int> temp;
686 temp[newtaxon] = con;
698 catch(exception& e) {
699 m->errorOut(e, "RemoveLineageCommand", "getTaxons");
703 /**************************************************************************************************/
704 string RemoveLineageCommand::removeConfidences(string tax) {
708 int taxLength = tax.length();
709 for(int i=0;i<taxLength;i++){
711 taxon = taxon.substr(0, taxon.find_first_of('(')); //rip off confidence
721 catch(exception& e) {
722 m->errorOut(e, "RemoveLineageCommand", "removeConfidences");
726 //**********************************************************************************************************************
727 //alignreport file has a column header line then all other lines contain 16 columns. we just want the first column since that contains the name
728 int RemoveLineageCommand::readAlign(){
730 string thisOutputDir = outputDir;
731 if (outputDir == "") { thisOutputDir += m->hasPath(alignfile); }
732 string outputFileName = thisOutputDir + m->getRootName(m->getSimpleName(alignfile)) + "pick.align.report";
735 m->openOutputFile(outputFileName, out);
738 m->openInputFile(alignfile, in);
741 bool wroteSomething = false;
743 //read column headers
744 for (int i = 0; i < 16; i++) {
745 if (!in.eof()) { in >> junk; out << junk << '\t'; }
751 if (m->control_pressed) { in.close(); out.close(); remove(outputFileName.c_str()); return 0; }
753 in >> name; //read from first column
755 //if this name is in the accnos file
756 if (names.count(name) == 0) {
757 wroteSomething = true;
762 for (int i = 0; i < 15; i++) {
763 if (!in.eof()) { in >> junk; out << junk << '\t'; }
768 }else {//still read just don't do anything with it
771 for (int i = 0; i < 15; i++) {
772 if (!in.eof()) { in >> junk; }
782 if (wroteSomething == false) { m->mothurOut("Your align file contains only sequences from " + taxons + "."); m->mothurOutEndLine(); }
783 outputNames.push_back(outputFileName); outputTypes["alignreport"].push_back(outputFileName);
788 catch(exception& e) {
789 m->errorOut(e, "RemoveLineageCommand", "readAlign");
793 //**********************************************************************************************************************