2 * getlineagecommand.cpp
5 * Created by westcott on 9/24/10.
6 * Copyright 2010 Schloss Lab. All rights reserved.
10 #include "getlineagecommand.h"
11 #include "sequence.hpp"
12 #include "listvector.hpp"
15 //**********************************************************************************************************************
16 vector<string> GetLineageCommand::getValidParameters(){
18 string Array[] = {"fasta","name", "group", "alignreport", "taxon", "dups", "list","taxonomy","outputdir","inputdir"};
19 vector<string> myArray (Array, Array+(sizeof(Array)/sizeof(string)));
23 m->errorOut(e, "GetLineageCommand", "getValidParameters");
27 //**********************************************************************************************************************
28 GetLineageCommand::GetLineageCommand(){
31 //initialize outputTypes
32 vector<string> tempOutNames;
33 outputTypes["fasta"] = tempOutNames;
34 outputTypes["taxonomy"] = tempOutNames;
35 outputTypes["name"] = tempOutNames;
36 outputTypes["group"] = tempOutNames;
37 outputTypes["alignreport"] = tempOutNames;
38 outputTypes["list"] = tempOutNames;
41 m->errorOut(e, "GetLineageCommand", "GetLineageCommand");
45 //**********************************************************************************************************************
46 vector<string> GetLineageCommand::getRequiredParameters(){
48 string Array[] = {"taxonomy","taxon"};
49 vector<string> myArray (Array, Array+(sizeof(Array)/sizeof(string)));
53 m->errorOut(e, "GetLineageCommand", "getRequiredParameters");
57 //**********************************************************************************************************************
58 vector<string> GetLineageCommand::getRequiredFiles(){
60 vector<string> myArray;
64 m->errorOut(e, "GetLineageCommand", "getRequiredFiles");
68 //**********************************************************************************************************************
69 GetLineageCommand::GetLineageCommand(string option) {
73 //allow user to run help
74 if(option == "help") { help(); abort = true; }
77 //valid paramters for this command
78 string Array[] = {"fasta","name", "group", "alignreport", "taxon", "dups", "list","taxonomy","outputdir","inputdir"};
79 vector<string> myArray (Array, Array+(sizeof(Array)/sizeof(string)));
81 OptionParser parser(option);
82 map<string,string> parameters = parser.getParameters();
84 ValidParameters validParameter;
85 map<string,string>::iterator it;
87 //check to make sure all parameters are valid for command
88 for (it = parameters.begin(); it != parameters.end(); it++) {
89 if (validParameter.isValidParameter(it->first, myArray, it->second) != true) { abort = true; }
92 //initialize outputTypes
93 vector<string> tempOutNames;
94 outputTypes["fasta"] = tempOutNames;
95 outputTypes["taxonomy"] = tempOutNames;
96 outputTypes["name"] = tempOutNames;
97 outputTypes["group"] = tempOutNames;
98 outputTypes["alignreport"] = tempOutNames;
99 outputTypes["list"] = tempOutNames;
101 //if the user changes the output directory command factory will send this info to us in the output parameter
102 outputDir = validParameter.validFile(parameters, "outputdir", false); if (outputDir == "not found"){ outputDir = ""; }
104 //if the user changes the input directory command factory will send this info to us in the output parameter
105 string inputDir = validParameter.validFile(parameters, "inputdir", false);
106 if (inputDir == "not found"){ inputDir = ""; }
109 it = parameters.find("alignreport");
110 //user has given a template file
111 if(it != parameters.end()){
112 path = m->hasPath(it->second);
113 //if the user has not given a path then, add inputdir. else leave path alone.
114 if (path == "") { parameters["alignreport"] = inputDir + it->second; }
117 it = parameters.find("fasta");
118 //user has given a template file
119 if(it != parameters.end()){
120 path = m->hasPath(it->second);
121 //if the user has not given a path then, add inputdir. else leave path alone.
122 if (path == "") { parameters["fasta"] = inputDir + it->second; }
125 it = parameters.find("list");
126 //user has given a template file
127 if(it != parameters.end()){
128 path = m->hasPath(it->second);
129 //if the user has not given a path then, add inputdir. else leave path alone.
130 if (path == "") { parameters["list"] = inputDir + it->second; }
133 it = parameters.find("name");
134 //user has given a template file
135 if(it != parameters.end()){
136 path = m->hasPath(it->second);
137 //if the user has not given a path then, add inputdir. else leave path alone.
138 if (path == "") { parameters["name"] = inputDir + it->second; }
141 it = parameters.find("group");
142 //user has given a template file
143 if(it != parameters.end()){
144 path = m->hasPath(it->second);
145 //if the user has not given a path then, add inputdir. else leave path alone.
146 if (path == "") { parameters["group"] = inputDir + it->second; }
149 it = parameters.find("taxonomy");
150 //user has given a template file
151 if(it != parameters.end()){
152 path = m->hasPath(it->second);
153 //if the user has not given a path then, add inputdir. else leave path alone.
154 if (path == "") { parameters["taxonomy"] = inputDir + it->second; }
159 //check for required parameters
160 fastafile = validParameter.validFile(parameters, "fasta", true);
161 if (fastafile == "not open") { abort = true; }
162 else if (fastafile == "not found") { fastafile = ""; }
164 namefile = validParameter.validFile(parameters, "name", true);
165 if (namefile == "not open") { abort = true; }
166 else if (namefile == "not found") { namefile = ""; }
168 groupfile = validParameter.validFile(parameters, "group", true);
169 if (groupfile == "not open") { abort = true; }
170 else if (groupfile == "not found") { groupfile = ""; }
172 alignfile = validParameter.validFile(parameters, "alignreport", true);
173 if (alignfile == "not open") { abort = true; }
174 else if (alignfile == "not found") { alignfile = ""; }
176 listfile = validParameter.validFile(parameters, "list", true);
177 if (listfile == "not open") { abort = true; }
178 else if (listfile == "not found") { listfile = ""; }
180 taxfile = validParameter.validFile(parameters, "taxonomy", true);
181 if (taxfile == "not open") { abort = true; }
182 else if (taxfile == "not found") { taxfile = ""; m->mothurOut("The taxonomy parameter is required for the get.lineage command."); m->mothurOutEndLine(); abort = true; }
184 string usedDups = "true";
185 string temp = validParameter.validFile(parameters, "dups", false);
186 if (temp == "not found") {
187 if (namefile != "") { temp = "true"; }
188 else { temp = "false"; usedDups = ""; }
190 dups = m->isTrue(temp);
192 taxons = validParameter.validFile(parameters, "taxon", false);
193 if (taxons == "not found") { taxons = ""; m->mothurOut("No taxons given, please correct."); m->mothurOutEndLine(); abort = true; }
196 if (taxons[0] == '\'') { taxons = taxons.substr(1); }
197 if (taxons[(taxons.length()-1)] == '\'') { taxons = taxons.substr(0, (taxons.length()-1)); }
201 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; }
203 if ((usedDups != "") && (namefile == "")) { m->mothurOut("You may only use dups with the name option."); m->mothurOutEndLine(); abort = true; }
208 catch(exception& e) {
209 m->errorOut(e, "GetLineageCommand", "GetLineageCommand");
213 //**********************************************************************************************************************
215 void GetLineageCommand::help(){
217 m->mothurOut("The get.lineage command reads a taxonomy file and any of the following file types: fasta, name, group, list or alignreport file.\n");
218 m->mothurOut("It outputs a file containing only the sequences from the taxonomy file that are from the taxon requested.\n");
219 m->mothurOut("The get.lineage command parameters are taxon, fasta, name, group, list, taxonomy, alignreport and dups. You must provide taxonomy and taxon.\n");
220 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");
221 m->mothurOut("The taxon parameter allows you to select the taxons you would like to get.\n");
222 m->mothurOut("You may enter your taxons with confidence scores, doing so will get only those sequences that belong to the taxonomy and whose cofidence scores is above the scores you give.\n");
223 m->mothurOut("If they belong to the taxonomy and have confidences below those you provide the sequence will not be selected.\n");
224 m->mothurOut("The get.lineage command should be in the following format: get.lineage(taxonomy=yourTaxonomyFile, taxon=yourTaxons).\n");
225 m->mothurOut("Example get.lineage(taxonomy=amazon.silva.taxonomy, taxon=Bacteria;Firmicutes;Bacilli;Lactobacillales;).\n");
226 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");
227 m->mothurOut("Example get.lineage(taxonomy=amazon.silva.taxonomy, taxon='Bacteria;Firmicutes;Bacilli;Lactobacillales;').\n");
228 m->mothurOut("Note: No spaces between parameter labels (i.e. fasta), '=' and parameters (i.e.yourFasta).\n\n");
230 catch(exception& e) {
231 m->errorOut(e, "GetLineageCommand", "help");
236 //**********************************************************************************************************************
238 int GetLineageCommand::execute(){
241 if (abort == true) { return 0; }
243 if (m->control_pressed) { return 0; }
245 //read through the correct file and output lines you want to keep
246 if (taxfile != "") { readTax(); } //fills the set of names to get
247 if (namefile != "") { readName(); }
248 if (fastafile != "") { readFasta(); }
249 if (groupfile != "") { readGroup(); }
250 if (alignfile != "") { readAlign(); }
251 if (listfile != "") { readList(); }
254 if (m->control_pressed) { outputTypes.clear(); for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } return 0; }
256 if (outputNames.size() != 0) {
257 m->mothurOutEndLine();
258 m->mothurOut("Output File Names: "); m->mothurOutEndLine();
259 for (int i = 0; i < outputNames.size(); i++) { m->mothurOut(outputNames[i]); m->mothurOutEndLine(); }
260 m->mothurOutEndLine();
266 catch(exception& e) {
267 m->errorOut(e, "GetLineageCommand", "execute");
272 //**********************************************************************************************************************
273 int GetLineageCommand::readFasta(){
275 string thisOutputDir = outputDir;
276 if (outputDir == "") { thisOutputDir += m->hasPath(fastafile); }
277 string outputFileName = thisOutputDir + m->getRootName(m->getSimpleName(fastafile)) + "pick" + m->getExtension(fastafile);
279 m->openOutputFile(outputFileName, out);
283 m->openInputFile(fastafile, in);
286 bool wroteSomething = false;
290 if (m->control_pressed) { in.close(); out.close(); remove(outputFileName.c_str()); return 0; }
292 Sequence currSeq(in);
293 name = currSeq.getName();
296 //if this name is in the accnos file
297 if (names.count(name) != 0) {
298 wroteSomething = true;
300 currSeq.printSequence(out);
308 if (wroteSomething == false) { m->mothurOut("Your file does not contain any sequence from the .accnos file."); m->mothurOutEndLine(); }
309 outputNames.push_back(outputFileName); outputTypes["fasta"].push_back(outputFileName);
314 catch(exception& e) {
315 m->errorOut(e, "GetLineageCommand", "readFasta");
319 //**********************************************************************************************************************
320 int GetLineageCommand::readList(){
322 string thisOutputDir = outputDir;
323 if (outputDir == "") { thisOutputDir += m->hasPath(listfile); }
324 string outputFileName = thisOutputDir + m->getRootName(m->getSimpleName(listfile)) + "pick" + m->getExtension(listfile);
326 m->openOutputFile(outputFileName, out);
329 m->openInputFile(listfile, in);
331 bool wroteSomething = false;
335 if (m->control_pressed) { in.close(); out.close(); remove(outputFileName.c_str()); return 0; }
337 //read in list vector
340 //make a new list vector
342 newList.setLabel(list.getLabel());
345 for (int i = 0; i < list.getNumBins(); i++) {
347 //parse out names that are in accnos file
348 string binnames = list.get(i);
350 string newNames = "";
351 while (binnames.find_first_of(',') != -1) {
352 string name = binnames.substr(0,binnames.find_first_of(','));
353 binnames = binnames.substr(binnames.find_first_of(',')+1, binnames.length());
355 //if that name is in the .accnos file, add it
356 if (names.count(name) != 0) { newNames += name + ","; }
360 if (names.count(binnames) != 0) { newNames += binnames + ","; }
362 //if there are names in this bin add to new list
363 if (newNames != "") {
364 newNames = newNames.substr(0, newNames.length()-1); //rip off extra comma
365 newList.push_back(newNames);
369 //print new listvector
370 if (newList.getNumBins() != 0) {
371 wroteSomething = true;
380 if (wroteSomething == false) { m->mothurOut("Your file does not contain any sequence from the .accnos file."); m->mothurOutEndLine(); }
381 outputNames.push_back(outputFileName); outputTypes["list"].push_back(outputFileName);
386 catch(exception& e) {
387 m->errorOut(e, "GetLineageCommand", "readList");
391 //**********************************************************************************************************************
392 int GetLineageCommand::readName(){
394 string thisOutputDir = outputDir;
395 if (outputDir == "") { thisOutputDir += m->hasPath(namefile); }
396 string outputFileName = thisOutputDir + m->getRootName(m->getSimpleName(namefile)) + "pick" + m->getExtension(namefile);
398 m->openOutputFile(outputFileName, out);
402 m->openInputFile(namefile, in);
403 string name, firstCol, secondCol;
405 bool wroteSomething = false;
410 if (m->control_pressed) { in.close(); out.close(); remove(outputFileName.c_str()); return 0; }
416 if (dups) { hold = secondCol; }
418 vector<string> parsedNames;
419 //parse second column saving each name
420 while (secondCol.find_first_of(',') != -1) {
421 name = secondCol.substr(0,secondCol.find_first_of(','));
422 secondCol = secondCol.substr(secondCol.find_first_of(',')+1, secondCol.length());
423 parsedNames.push_back(name);
426 //get name after last ,
427 parsedNames.push_back(secondCol);
429 vector<string> validSecond;
430 for (int i = 0; i < parsedNames.size(); i++) {
431 if (names.count(parsedNames[i]) != 0) {
432 validSecond.push_back(parsedNames[i]);
436 if ((dups) && (validSecond.size() != 0)) { //dups = true and we want to add someone, then add everyone
437 for (int i = 0; i < parsedNames.size(); i++) { names.insert(parsedNames[i]); }
438 out << firstCol << '\t' << hold << endl;
439 wroteSomething = true;
441 //if the name in the first column is in the set then print it and any other names in second column also in set
442 if (names.count(firstCol) != 0) {
444 wroteSomething = true;
446 out << firstCol << '\t';
448 //you know you have at least one valid second since first column is valid
449 for (int i = 0; i < validSecond.size()-1; i++) { out << validSecond[i] << ','; }
450 out << validSecond[validSecond.size()-1] << endl;
453 //make first name in set you come to first column and then add the remaining names to second column
455 //you want part of this row
456 if (validSecond.size() != 0) {
458 wroteSomething = true;
460 out << validSecond[0] << '\t';
462 //you know you have at least one valid second since first column is valid
463 for (int i = 0; i < validSecond.size()-1; i++) { out << validSecond[i] << ','; }
464 out << validSecond[validSecond.size()-1] << endl;
473 if (wroteSomething == false) { m->mothurOut("Your file does not contain any sequence from the .accnos file."); m->mothurOutEndLine(); }
474 outputNames.push_back(outputFileName); outputTypes["name"].push_back(outputFileName);
479 catch(exception& e) {
480 m->errorOut(e, "GetLineageCommand", "readName");
485 //**********************************************************************************************************************
486 int GetLineageCommand::readGroup(){
488 string thisOutputDir = outputDir;
489 if (outputDir == "") { thisOutputDir += m->hasPath(groupfile); }
490 string outputFileName = thisOutputDir + m->getRootName(m->getSimpleName(groupfile)) + "pick" + m->getExtension(groupfile);
492 m->openOutputFile(outputFileName, out);
496 m->openInputFile(groupfile, in);
499 bool wroteSomething = false;
503 if (m->control_pressed) { in.close(); out.close(); remove(outputFileName.c_str()); return 0; }
506 in >> name; //read from first column
507 in >> group; //read from second column
509 //if this name is in the accnos file
510 if (names.count(name) != 0) {
511 wroteSomething = true;
513 out << name << '\t' << group << endl;
521 if (wroteSomething == false) { m->mothurOut("Your file does not contain any sequence from the .accnos file."); m->mothurOutEndLine(); }
522 outputNames.push_back(outputFileName); outputTypes["group"].push_back(outputFileName);
527 catch(exception& e) {
528 m->errorOut(e, "GetLineageCommand", "readGroup");
532 //**********************************************************************************************************************
533 int GetLineageCommand::readTax(){
535 string thisOutputDir = outputDir;
536 if (outputDir == "") { thisOutputDir += m->hasPath(taxfile); }
537 string outputFileName = thisOutputDir + m->getRootName(m->getSimpleName(taxfile)) + "pick" + m->getExtension(taxfile);
539 m->openOutputFile(outputFileName, out);
542 m->openInputFile(taxfile, in);
545 //bool wroteSomething = false;
547 bool taxonsHasConfidence = false;
548 vector< map<string, int> > searchTaxons;
549 string noConfidenceTaxons = taxons;
550 int hasConPos = taxons.find_first_of('(');
551 if (hasConPos != string::npos) {
552 taxonsHasConfidence = true;
553 searchTaxons = getTaxons(taxons);
554 noConfidenceTaxons = removeConfidences(taxons);
560 if (m->control_pressed) { in.close(); out.close(); remove(outputFileName.c_str()); return 0; }
562 in >> name; //read from first column
563 in >> tax; //read from second column
568 //if the users file contains confidence scores we want to ignore them when searching for the taxons, unless the taxon has them
569 if (!taxonsHasConfidence) {
570 int hasConfidences = tax.find_first_of('(');
571 if (hasConfidences != string::npos) {
572 newtax = removeConfidences(tax);
575 int pos = newtax.find(taxons);
577 if (pos != string::npos) { //this sequence contains the taxon the user wants
579 out << name << '\t' << tax << endl;
582 }else{//if taxons has them and you don't them remove taxons
583 int hasConfidences = tax.find_first_of('(');
584 if (hasConfidences == string::npos) {
586 int pos = newtax.find(noConfidenceTaxons);
588 if (pos != string::npos) { //this sequence contains the taxon the user wants
590 out << name << '\t' << tax << endl;
592 }else { //both have confidences so we want to make sure the users confidences are greater then or equal to the taxons
593 //first remove confidences from both and see if the taxonomy exists
595 string noNewTax = tax;
596 int hasConfidences = tax.find_first_of('(');
597 if (hasConfidences != string::npos) {
598 noNewTax = removeConfidences(tax);
601 int pos = noNewTax.find(noConfidenceTaxons);
603 if (pos != string::npos) { //if yes, then are the confidences okay
606 vector< map<string, int> > usersTaxon = getTaxons(newtax);
608 //the usersTaxon is most likely longer than the searchTaxons, and searchTaxon[0] may relate to userTaxon[4]
609 //we want to "line them up", so we will find the the index where the searchstring starts
611 for (int i = 0; i < usersTaxon.size(); i++) {
613 if (usersTaxon[i].begin()->first == searchTaxons[0].begin()->first) {
616 bool goodspot = true;
617 //is this really the start, or are we dealing with a taxon of the same name?
618 while ((spot < searchTaxons.size()) && ((i+spot) < usersTaxon.size())) {
619 if (usersTaxon[i+spot].begin()->first != searchTaxons[spot].begin()->first) { goodspot = false; break; }
623 if (goodspot) { break; }
627 for (int i = 0; i < searchTaxons.size(); i++) {
629 if ((i+index) < usersTaxon.size()) { //just in case, should never be false
630 if (usersTaxon[i+index].begin()->second < searchTaxons[i].begin()->second) { //is the users cutoff less than the search taxons
640 //passed the test so add you
643 out << name << '\t' << tax << endl;
656 if (names.size() == 0) { m->mothurOut("Your taxonomy file does not contain any sequences from " + taxons + "."); m->mothurOutEndLine(); }
657 outputNames.push_back(outputFileName); outputTypes["taxonomy"].push_back(outputFileName);
662 catch(exception& e) {
663 m->errorOut(e, "GetLineageCommand", "readTax");
667 /**************************************************************************************************/
668 vector< map<string, int> > GetLineageCommand::getTaxons(string tax) {
671 vector< map<string, int> > t;
673 int taxLength = tax.length();
674 for(int i=0;i<taxLength;i++){
677 int openParen = taxon.find_first_of('(');
678 int closeParen = taxon.find_last_of(')');
680 string newtaxon = taxon.substr(0, openParen); //rip off confidence
681 string 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, "GetLineageCommand", "getTaxons");
703 /**************************************************************************************************/
704 string GetLineageCommand::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, "GetLineageCommand", "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 GetLineageCommand::readAlign(){
730 string thisOutputDir = outputDir;
731 if (outputDir == "") { thisOutputDir += m->hasPath(alignfile); }
732 string outputFileName = thisOutputDir + m->getRootName(m->getSimpleName(alignfile)) + "pick.align.report";
734 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'; }
752 if (m->control_pressed) { in.close(); out.close(); remove(outputFileName.c_str()); return 0; }
755 in >> name; //read from first column
757 //if this name is in the accnos file
758 if (names.count(name) != 0) {
759 wroteSomething = true;
764 for (int i = 0; i < 15; i++) {
765 if (!in.eof()) { in >> junk; out << junk << '\t'; }
770 }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 file does not contain any sequence from the .accnos file."); m->mothurOutEndLine(); }
784 outputNames.push_back(outputFileName); outputTypes["alignreport"].push_back(outputFileName);
789 catch(exception& e) {
790 m->errorOut(e, "GetLineageCommand", "readAlign");
794 //**********************************************************************************************************************