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(){
30 //initialize outputTypes
31 vector<string> tempOutNames;
32 outputTypes["fasta"] = tempOutNames;
33 outputTypes["taxonomy"] = tempOutNames;
34 outputTypes["name"] = tempOutNames;
35 outputTypes["group"] = tempOutNames;
36 outputTypes["alignreport"] = tempOutNames;
37 outputTypes["list"] = tempOutNames;
40 m->errorOut(e, "RemoveLineageCommand", "RemoveLineageCommand");
44 //**********************************************************************************************************************
45 vector<string> RemoveLineageCommand::getRequiredParameters(){
47 string Array[] = {"taxonomy"};
48 vector<string> myArray (Array, Array+(sizeof(Array)/sizeof(string)));
52 m->errorOut(e, "RemoveLineageCommand", "getRequiredParameters");
56 //**********************************************************************************************************************
57 vector<string> RemoveLineageCommand::getRequiredFiles(){
59 vector<string> myArray;
63 m->errorOut(e, "RemoveLineageCommand", "getRequiredFiles");
67 //**********************************************************************************************************************
68 RemoveLineageCommand::RemoveLineageCommand(string option) {
72 //allow user to run help
73 if(option == "help") { help(); abort = true; }
76 //valid paramters for this command
77 string Array[] = {"fasta","name", "group", "alignreport", "taxon", "dups", "list","taxonomy","outputdir","inputdir"};
78 vector<string> myArray (Array, Array+(sizeof(Array)/sizeof(string)));
80 OptionParser parser(option);
81 map<string,string> parameters = parser.getParameters();
83 ValidParameters validParameter;
84 map<string,string>::iterator it;
86 //check to make sure all parameters are valid for command
87 for (it = parameters.begin(); it != parameters.end(); it++) {
88 if (validParameter.isValidParameter(it->first, myArray, it->second) != true) { abort = true; }
91 //initialize outputTypes
92 vector<string> tempOutNames;
93 outputTypes["fasta"] = tempOutNames;
94 outputTypes["taxonomy"] = tempOutNames;
95 outputTypes["name"] = tempOutNames;
96 outputTypes["group"] = tempOutNames;
97 outputTypes["alignreport"] = tempOutNames;
98 outputTypes["list"] = tempOutNames;
100 //if the user changes the output directory command factory will send this info to us in the output parameter
101 outputDir = validParameter.validFile(parameters, "outputdir", false); if (outputDir == "not found"){ outputDir = ""; }
103 //if the user changes the input directory command factory will send this info to us in the output parameter
104 string inputDir = validParameter.validFile(parameters, "inputdir", false);
105 if (inputDir == "not found"){ inputDir = ""; }
108 it = parameters.find("alignreport");
109 //user has given a template file
110 if(it != parameters.end()){
111 path = m->hasPath(it->second);
112 //if the user has not given a path then, add inputdir. else leave path alone.
113 if (path == "") { parameters["alignreport"] = inputDir + it->second; }
116 it = parameters.find("fasta");
117 //user has given a template file
118 if(it != parameters.end()){
119 path = m->hasPath(it->second);
120 //if the user has not given a path then, add inputdir. else leave path alone.
121 if (path == "") { parameters["fasta"] = inputDir + it->second; }
124 it = parameters.find("list");
125 //user has given a template file
126 if(it != parameters.end()){
127 path = m->hasPath(it->second);
128 //if the user has not given a path then, add inputdir. else leave path alone.
129 if (path == "") { parameters["list"] = inputDir + it->second; }
132 it = parameters.find("name");
133 //user has given a template file
134 if(it != parameters.end()){
135 path = m->hasPath(it->second);
136 //if the user has not given a path then, add inputdir. else leave path alone.
137 if (path == "") { parameters["name"] = inputDir + it->second; }
140 it = parameters.find("group");
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["group"] = inputDir + it->second; }
148 it = parameters.find("taxonomy");
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["taxonomy"] = inputDir + it->second; }
158 //check for required parameters
159 fastafile = validParameter.validFile(parameters, "fasta", true);
160 if (fastafile == "not open") { abort = true; }
161 else if (fastafile == "not found") { fastafile = ""; }
163 namefile = validParameter.validFile(parameters, "name", true);
164 if (namefile == "not open") { abort = true; }
165 else if (namefile == "not found") { namefile = ""; }
167 groupfile = validParameter.validFile(parameters, "group", true);
168 if (groupfile == "not open") { abort = true; }
169 else if (groupfile == "not found") { groupfile = ""; }
171 alignfile = validParameter.validFile(parameters, "alignreport", true);
172 if (alignfile == "not open") { abort = true; }
173 else if (alignfile == "not found") { alignfile = ""; }
175 listfile = validParameter.validFile(parameters, "list", true);
176 if (listfile == "not open") { abort = true; }
177 else if (listfile == "not found") { listfile = ""; }
179 taxfile = validParameter.validFile(parameters, "taxonomy", true);
180 if (taxfile == "not open") { abort = true; }
181 else if (taxfile == "not found") { taxfile = ""; m->mothurOut("The taxonomy parameter is required for the get.lineage command."); m->mothurOutEndLine(); abort = true; }
183 string usedDups = "true";
184 string temp = validParameter.validFile(parameters, "dups", false);
185 if (temp == "not found") {
186 if (namefile != "") { temp = "true"; }
187 else { temp = "false"; usedDups = ""; }
189 dups = m->isTrue(temp);
191 taxons = validParameter.validFile(parameters, "taxon", false);
192 if (taxons == "not found") { taxons = ""; m->mothurOut("No taxons given, please correct."); m->mothurOutEndLine(); abort = true; }
195 if (taxons[0] == '\'') { taxons = taxons.substr(1); }
196 if (taxons[(taxons.length()-1)] == '\'') { taxons = taxons.substr(0, (taxons.length()-1)); }
200 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; }
202 if ((usedDups != "") && (namefile == "")) { m->mothurOut("You may only use dups with the name option."); m->mothurOutEndLine(); abort = true; }
207 catch(exception& e) {
208 m->errorOut(e, "RemoveLineageCommand", "RemoveLineageCommand");
212 //**********************************************************************************************************************
214 void RemoveLineageCommand::help(){
216 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");
217 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");
218 m->mothurOut("The remove.lineage command parameters are taxon, fasta, name, group, list, taxonomy, alignreport and dups. You must provide taxonomy and taxon.\n");
219 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");
220 m->mothurOut("The taxon parameter allows you to select the taxons you would like to remove.\n");
221 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");
222 m->mothurOut("If they belong to the taxonomy and have confidences above those you provide the sequence will not be removed.\n");
223 m->mothurOut("The remove.lineage command should be in the following format: remove.lineage(taxonomy=yourTaxonomyFile, taxon=yourTaxons).\n");
224 m->mothurOut("Example remove.lineage(taxonomy=amazon.silva.taxonomy, taxon=Bacteria;Firmicutes;Bacilli;Lactobacillales;).\n");
225 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");
226 m->mothurOut("Example remove.lineage(taxonomy=amazon.silva.taxonomy, taxon='Bacteria;Firmicutes;Bacilli;Lactobacillales;').\n");
227 m->mothurOut("Note: No spaces between parameter labels (i.e. fasta), '=' and parameters (i.e.yourFasta).\n\n");
229 catch(exception& e) {
230 m->errorOut(e, "RemoveLineageCommand", "help");
235 //**********************************************************************************************************************
237 int RemoveLineageCommand::execute(){
240 if (abort == true) { return 0; }
242 if (m->control_pressed) { return 0; }
244 //read through the correct file and output lines you want to keep
245 if (taxfile != "") { readTax(); } //fills the set of names to remove
246 if (namefile != "") { readName(); }
247 if (fastafile != "") { readFasta(); }
248 if (groupfile != "") { readGroup(); }
249 if (alignfile != "") { readAlign(); }
250 if (listfile != "") { readList(); }
253 if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } return 0; }
255 if (outputNames.size() != 0) {
256 m->mothurOutEndLine();
257 m->mothurOut("Output File Names: "); m->mothurOutEndLine();
258 for (int i = 0; i < outputNames.size(); i++) { m->mothurOut(outputNames[i]); m->mothurOutEndLine(); }
259 m->mothurOutEndLine();
265 catch(exception& e) {
266 m->errorOut(e, "RemoveLineageCommand", "execute");
271 //**********************************************************************************************************************
272 int RemoveLineageCommand::readFasta(){
274 string thisOutputDir = outputDir;
275 if (outputDir == "") { thisOutputDir += m->hasPath(fastafile); }
276 string outputFileName = thisOutputDir + m->getRootName(m->getSimpleName(fastafile)) + "pick" + m->getExtension(fastafile);
279 m->openOutputFile(outputFileName, out);
282 m->openInputFile(fastafile, in);
285 bool wroteSomething = false;
288 if (m->control_pressed) { in.close(); out.close(); remove(outputFileName.c_str()); return 0; }
290 Sequence currSeq(in);
291 name = currSeq.getName();
294 //if this name is in the accnos file
295 if (names.count(name) == 0) {
296 wroteSomething = true;
298 currSeq.printSequence(out);
306 if (wroteSomething == false) { m->mothurOut("Your fasta file contains only sequences from " + taxons + "."); m->mothurOutEndLine(); }
307 outputNames.push_back(outputFileName); outputTypes["fasta"].push_back(outputFileName);
312 catch(exception& e) {
313 m->errorOut(e, "RemoveLineageCommand", "readFasta");
317 //**********************************************************************************************************************
318 int RemoveLineageCommand::readList(){
320 string thisOutputDir = outputDir;
321 if (outputDir == "") { thisOutputDir += m->hasPath(listfile); }
322 string outputFileName = thisOutputDir + m->getRootName(m->getSimpleName(listfile)) + "pick" + m->getExtension(listfile);
325 m->openOutputFile(outputFileName, out);
328 m->openInputFile(listfile, in);
330 bool wroteSomething = false;
333 //read in list vector
336 //make a new list vector
338 newList.setLabel(list.getLabel());
341 for (int i = 0; i < list.getNumBins(); i++) {
342 if (m->control_pressed) { in.close(); out.close(); remove(outputFileName.c_str()); return 0; }
344 //parse out names that are in accnos file
345 string binnames = list.get(i);
347 string newNames = "";
348 while (binnames.find_first_of(',') != -1) {
349 string name = binnames.substr(0,binnames.find_first_of(','));
350 binnames = binnames.substr(binnames.find_first_of(',')+1, binnames.length());
352 //if that name is in the .accnos file, add it
353 if (names.count(name) == 0) { newNames += name + ","; }
357 if (names.count(binnames) == 0) { newNames += binnames + ","; }
359 //if there are names in this bin add to new list
360 if (newNames != "") {
361 newNames = newNames.substr(0, newNames.length()-1); //rip off extra comma
362 newList.push_back(newNames);
366 //print new listvector
367 if (newList.getNumBins() != 0) {
368 wroteSomething = true;
377 if (wroteSomething == false) { m->mothurOut("Your list file contains only sequences from " + taxons + "."); m->mothurOutEndLine(); }
378 outputNames.push_back(outputFileName); outputTypes["list"].push_back(outputFileName);
383 catch(exception& e) {
384 m->errorOut(e, "RemoveLineageCommand", "readList");
388 //**********************************************************************************************************************
389 int RemoveLineageCommand::readName(){
391 string thisOutputDir = outputDir;
392 if (outputDir == "") { thisOutputDir += m->hasPath(namefile); }
393 string outputFileName = thisOutputDir + m->getRootName(m->getSimpleName(namefile)) + "pick" + m->getExtension(namefile);
396 m->openOutputFile(outputFileName, out);
399 m->openInputFile(namefile, in);
400 string name, firstCol, secondCol;
402 bool wroteSomething = false;
405 if (m->control_pressed) { in.close(); out.close(); remove(outputFileName.c_str()); return 0; }
410 vector<string> parsedNames;
411 //parse second column saving each name
412 while (secondCol.find_first_of(',') != -1) {
413 name = secondCol.substr(0,secondCol.find_first_of(','));
414 secondCol = secondCol.substr(secondCol.find_first_of(',')+1, secondCol.length());
415 parsedNames.push_back(name);
418 //get name after last ,
419 parsedNames.push_back(secondCol);
421 vector<string> validSecond; validSecond.clear();
422 for (int i = 0; i < parsedNames.size(); i++) {
423 if (names.count(parsedNames[i]) == 0) {
424 validSecond.push_back(parsedNames[i]);
428 if ((dups) && (validSecond.size() != parsedNames.size())) { //if dups is true and we want to get rid of anyone, get rid of everyone
429 for (int i = 0; i < parsedNames.size(); i++) { names.insert(parsedNames[i]); }
431 //if the name in the first column is in the set then print it and any other names in second column also in set
432 if (names.count(firstCol) == 0) {
434 wroteSomething = true;
436 out << firstCol << '\t';
438 //you know you have at least one valid second since first column is valid
439 for (int i = 0; i < validSecond.size()-1; i++) { out << validSecond[i] << ','; }
440 out << validSecond[validSecond.size()-1] << endl;
442 //make first name in set you come to first column and then add the remaining names to second column
445 //you want part of this row
446 if (validSecond.size() != 0) {
448 wroteSomething = true;
450 out << validSecond[0] << '\t';
452 //you know you have at least one valid second since first column is valid
453 for (int i = 0; i < validSecond.size()-1; i++) { out << validSecond[i] << ','; }
454 out << validSecond[validSecond.size()-1] << endl;
463 if (wroteSomething == false) { m->mothurOut("Your name file contains only sequences from " + taxons + "."); m->mothurOutEndLine(); }
464 outputNames.push_back(outputFileName); outputTypes["name"].push_back(outputFileName);
468 catch(exception& e) {
469 m->errorOut(e, "RemoveLineageCommand", "readName");
474 //**********************************************************************************************************************
475 int RemoveLineageCommand::readGroup(){
477 string thisOutputDir = outputDir;
478 if (outputDir == "") { thisOutputDir += m->hasPath(groupfile); }
479 string outputFileName = thisOutputDir + m->getRootName(m->getSimpleName(groupfile)) + "pick" + m->getExtension(groupfile);
482 m->openOutputFile(outputFileName, out);
485 m->openInputFile(groupfile, in);
488 bool wroteSomething = false;
491 if (m->control_pressed) { in.close(); out.close(); remove(outputFileName.c_str()); return 0; }
493 in >> name; //read from first column
494 in >> group; //read from second column
496 //if this name is in the accnos file
497 if (names.count(name) == 0) {
498 wroteSomething = true;
499 out << name << '\t' << group << endl;
507 if (wroteSomething == false) { m->mothurOut("Your group file contains only sequences from " + taxons + "."); m->mothurOutEndLine(); }
508 outputNames.push_back(outputFileName); outputTypes["group"].push_back(outputFileName);
512 catch(exception& e) {
513 m->errorOut(e, "RemoveLineageCommand", "readGroup");
517 //**********************************************************************************************************************
518 int RemoveLineageCommand::readTax(){
520 string thisOutputDir = outputDir;
521 if (outputDir == "") { thisOutputDir += m->hasPath(taxfile); }
522 string outputFileName = thisOutputDir + m->getRootName(m->getSimpleName(taxfile)) + "pick" + m->getExtension(taxfile);
524 m->openOutputFile(outputFileName, out);
527 m->openInputFile(taxfile, in);
530 bool wroteSomething = false;
532 bool taxonsHasConfidence = false;
533 vector< map<string, int> > searchTaxons;
534 string noConfidenceTaxons = taxons;
535 int hasConPos = taxons.find_first_of('(');
536 if (hasConPos != string::npos) {
537 taxonsHasConfidence = true;
538 searchTaxons = getTaxons(taxons);
539 noConfidenceTaxons = removeConfidences(taxons);
545 if (m->control_pressed) { in.close(); out.close(); remove(outputFileName.c_str()); return 0; }
547 in >> name; //read from first column
548 in >> tax; //read from second column
552 //if the users file contains confidence scores we want to ignore them when searching for the taxons, unless the taxon has them
553 if (!taxonsHasConfidence) {
555 int hasConfidences = tax.find_first_of('(');
556 if (hasConfidences != string::npos) {
557 newtax = removeConfidences(tax);
560 int pos = newtax.find(taxons);
562 if (pos == string::npos) {
563 wroteSomething = true;
564 out << name << '\t' << tax << endl;
565 }else{ //this sequence contains the taxon the user wants to remove
569 }else{//if taxons has them and you don't them remove taxons
570 int hasConfidences = tax.find_first_of('(');
571 if (hasConfidences == string::npos) {
573 int pos = newtax.find(noConfidenceTaxons);
575 if (pos == string::npos) {
576 wroteSomething = true;
577 out << name << '\t' << tax << endl;
578 }else{ //this sequence contains the taxon the user wants to remove
581 }else { //both have confidences so we want to make sure the users confidences are greater then or equal to the taxons
582 //first remove confidences from both and see if the taxonomy exists
584 string noNewTax = tax;
585 int hasConfidences = tax.find_first_of('(');
586 if (hasConfidences != string::npos) {
587 noNewTax = removeConfidences(tax);
590 int pos = noNewTax.find(noConfidenceTaxons);
592 if (pos != string::npos) { //if yes, then are the confidences okay
595 vector< map<string, int> > usersTaxon = getTaxons(newtax);
597 //the usersTaxon is most likely longer than the searchTaxons, and searchTaxon[0] may relate to userTaxon[4]
598 //we want to "line them up", so we will find the the index where the searchstring starts
600 for (int i = 0; i < usersTaxon.size(); i++) {
602 if (usersTaxon[i].begin()->first == searchTaxons[0].begin()->first) {
605 bool goodspot = true;
606 //is this really the start, or are we dealing with a taxon of the same name?
607 while ((spot < searchTaxons.size()) && ((i+spot) < usersTaxon.size())) {
608 if (usersTaxon[i+spot].begin()->first != searchTaxons[spot].begin()->first) { goodspot = false; break; }
612 if (goodspot) { break; }
616 for (int i = 0; i < searchTaxons.size(); i++) {
618 if ((i+index) < usersTaxon.size()) { //just in case, should never be false
619 if (usersTaxon[i+index].begin()->second < searchTaxons[i].begin()->second) { //is the users cutoff less than the search taxons
629 //passed the test so remove you
633 wroteSomething = true;
634 out << name << '\t' << tax << endl;
637 wroteSomething = true;
638 out << name << '\t' << tax << endl;
651 if (!wroteSomething) { m->mothurOut("Your taxonomy file contains only sequences from " + taxons + "."); m->mothurOutEndLine(); }
652 outputNames.push_back(outputFileName); outputTypes["taxonomy"].push_back(outputFileName);
657 catch(exception& e) {
658 m->errorOut(e, "RemoveLineageCommand", "readTax");
662 /**************************************************************************************************/
663 vector< map<string, int> > RemoveLineageCommand::getTaxons(string tax) {
666 vector< map<string, int> > t;
668 int taxLength = tax.length();
669 for(int i=0;i<taxLength;i++){
672 int openParen = taxon.find_first_of('(');
673 int closeParen = taxon.find_last_of(')');
675 string newtaxon, confidence;
676 if ((openParen != string::npos) && (closeParen != string::npos)) {
677 newtaxon = taxon.substr(0, openParen); //rip off confidence
678 confidence = taxon.substr((openParen+1), (closeParen-openParen-1));
684 convert(confidence, con);
686 map<string, int> temp;
687 temp[newtaxon] = con;
699 catch(exception& e) {
700 m->errorOut(e, "RemoveLineageCommand", "getTaxons");
704 /**************************************************************************************************/
705 string RemoveLineageCommand::removeConfidences(string tax) {
709 int taxLength = tax.length();
710 for(int i=0;i<taxLength;i++){
712 taxon = taxon.substr(0, taxon.find_first_of('(')); //rip off confidence
722 catch(exception& e) {
723 m->errorOut(e, "RemoveLineageCommand", "removeConfidences");
727 //**********************************************************************************************************************
728 //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
729 int RemoveLineageCommand::readAlign(){
731 string thisOutputDir = outputDir;
732 if (outputDir == "") { thisOutputDir += m->hasPath(alignfile); }
733 string outputFileName = thisOutputDir + m->getRootName(m->getSimpleName(alignfile)) + "pick.align.report";
736 m->openOutputFile(outputFileName, out);
739 m->openInputFile(alignfile, in);
742 bool wroteSomething = false;
744 //read column headers
745 for (int i = 0; i < 16; i++) {
746 if (!in.eof()) { in >> junk; out << junk << '\t'; }
752 if (m->control_pressed) { in.close(); out.close(); remove(outputFileName.c_str()); return 0; }
754 in >> name; //read from first column
756 //if this name is in the accnos file
757 if (names.count(name) == 0) {
758 wroteSomething = true;
763 for (int i = 0; i < 15; i++) {
764 if (!in.eof()) { in >> junk; out << junk << '\t'; }
769 }else {//still read just don't do anything with it
772 for (int i = 0; i < 15; i++) {
773 if (!in.eof()) { in >> junk; }
783 if (wroteSomething == false) { m->mothurOut("Your align file contains only sequences from " + taxons + "."); m->mothurOutEndLine(); }
784 outputNames.push_back(outputFileName); outputTypes["alignreport"].push_back(outputFileName);
789 catch(exception& e) {
790 m->errorOut(e, "RemoveLineageCommand", "readAlign");
794 //**********************************************************************************************************************