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(){
30 abort = true; calledHelp = true;
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, "GetLineageCommand", "GetLineageCommand");
44 //**********************************************************************************************************************
45 vector<string> GetLineageCommand::getRequiredParameters(){
47 string Array[] = {"taxonomy","taxon"};
48 vector<string> myArray (Array, Array+(sizeof(Array)/sizeof(string)));
52 m->errorOut(e, "GetLineageCommand", "getRequiredParameters");
56 //**********************************************************************************************************************
57 vector<string> GetLineageCommand::getRequiredFiles(){
59 vector<string> myArray;
63 m->errorOut(e, "GetLineageCommand", "getRequiredFiles");
67 //**********************************************************************************************************************
68 GetLineageCommand::GetLineageCommand(string option) {
70 abort = false; calledHelp = false;
72 //allow user to run help
73 if(option == "help") { help(); abort = true; calledHelp = 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, "GetLineageCommand", "GetLineageCommand");
212 //**********************************************************************************************************************
214 void GetLineageCommand::help(){
216 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");
217 m->mothurOut("It outputs a file containing only the sequences from the taxonomy file that are from the taxon requested.\n");
218 m->mothurOut("The get.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 get.\n");
221 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");
222 m->mothurOut("If they belong to the taxonomy and have confidences below those you provide the sequence will not be selected.\n");
223 m->mothurOut("The get.lineage command should be in the following format: get.lineage(taxonomy=yourTaxonomyFile, taxon=yourTaxons).\n");
224 m->mothurOut("Example get.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 get.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, "GetLineageCommand", "help");
235 //**********************************************************************************************************************
237 int GetLineageCommand::execute(){
240 if (abort == true) { if (calledHelp) { return 0; } return 2; }
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 get
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) { outputTypes.clear(); 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, "GetLineageCommand", "execute");
271 //**********************************************************************************************************************
272 int GetLineageCommand::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);
278 m->openOutputFile(outputFileName, out);
282 m->openInputFile(fastafile, in);
285 bool wroteSomething = false;
289 if (m->control_pressed) { in.close(); out.close(); remove(outputFileName.c_str()); return 0; }
291 Sequence currSeq(in);
292 name = currSeq.getName();
295 //if this name is in the accnos file
296 if (names.count(name) != 0) {
297 wroteSomething = true;
299 currSeq.printSequence(out);
307 if (wroteSomething == false) { m->mothurOut("Your file does not contain any sequence from the .accnos file."); m->mothurOutEndLine(); }
308 outputNames.push_back(outputFileName); outputTypes["fasta"].push_back(outputFileName);
313 catch(exception& e) {
314 m->errorOut(e, "GetLineageCommand", "readFasta");
318 //**********************************************************************************************************************
319 int GetLineageCommand::readList(){
321 string thisOutputDir = outputDir;
322 if (outputDir == "") { thisOutputDir += m->hasPath(listfile); }
323 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;
334 if (m->control_pressed) { in.close(); out.close(); remove(outputFileName.c_str()); return 0; }
336 //read in list vector
339 //make a new list vector
341 newList.setLabel(list.getLabel());
344 for (int i = 0; i < list.getNumBins(); i++) {
346 //parse out names that are in accnos file
347 string binnames = list.get(i);
349 string newNames = "";
350 while (binnames.find_first_of(',') != -1) {
351 string name = binnames.substr(0,binnames.find_first_of(','));
352 binnames = binnames.substr(binnames.find_first_of(',')+1, binnames.length());
354 //if that name is in the .accnos file, add it
355 if (names.count(name) != 0) { newNames += name + ","; }
359 if (names.count(binnames) != 0) { newNames += binnames + ","; }
361 //if there are names in this bin add to new list
362 if (newNames != "") {
363 newNames = newNames.substr(0, newNames.length()-1); //rip off extra comma
364 newList.push_back(newNames);
368 //print new listvector
369 if (newList.getNumBins() != 0) {
370 wroteSomething = true;
379 if (wroteSomething == false) { m->mothurOut("Your file does not contain any sequence from the .accnos file."); m->mothurOutEndLine(); }
380 outputNames.push_back(outputFileName); outputTypes["list"].push_back(outputFileName);
385 catch(exception& e) {
386 m->errorOut(e, "GetLineageCommand", "readList");
390 //**********************************************************************************************************************
391 int GetLineageCommand::readName(){
393 string thisOutputDir = outputDir;
394 if (outputDir == "") { thisOutputDir += m->hasPath(namefile); }
395 string outputFileName = thisOutputDir + m->getRootName(m->getSimpleName(namefile)) + "pick" + m->getExtension(namefile);
397 m->openOutputFile(outputFileName, out);
401 m->openInputFile(namefile, in);
402 string name, firstCol, secondCol;
404 bool wroteSomething = false;
409 if (m->control_pressed) { in.close(); out.close(); remove(outputFileName.c_str()); return 0; }
415 if (dups) { hold = secondCol; }
417 vector<string> parsedNames;
418 //parse second column saving each name
419 while (secondCol.find_first_of(',') != -1) {
420 name = secondCol.substr(0,secondCol.find_first_of(','));
421 secondCol = secondCol.substr(secondCol.find_first_of(',')+1, secondCol.length());
422 parsedNames.push_back(name);
425 //get name after last ,
426 parsedNames.push_back(secondCol);
428 vector<string> validSecond;
429 for (int i = 0; i < parsedNames.size(); i++) {
430 if (names.count(parsedNames[i]) != 0) {
431 validSecond.push_back(parsedNames[i]);
435 if ((dups) && (validSecond.size() != 0)) { //dups = true and we want to add someone, then add everyone
436 for (int i = 0; i < parsedNames.size(); i++) { names.insert(parsedNames[i]); }
437 out << firstCol << '\t' << hold << endl;
438 wroteSomething = true;
440 //if the name in the first column is in the set then print it and any other names in second column also in set
441 if (names.count(firstCol) != 0) {
443 wroteSomething = true;
445 out << firstCol << '\t';
447 //you know you have at least one valid second since first column is valid
448 for (int i = 0; i < validSecond.size()-1; i++) { out << validSecond[i] << ','; }
449 out << validSecond[validSecond.size()-1] << endl;
452 //make first name in set you come to first column and then add the remaining names to second column
454 //you want part of this row
455 if (validSecond.size() != 0) {
457 wroteSomething = true;
459 out << validSecond[0] << '\t';
461 //you know you have at least one valid second since first column is valid
462 for (int i = 0; i < validSecond.size()-1; i++) { out << validSecond[i] << ','; }
463 out << validSecond[validSecond.size()-1] << endl;
472 if (wroteSomething == false) { m->mothurOut("Your file does not contain any sequence from the .accnos file."); m->mothurOutEndLine(); }
473 outputNames.push_back(outputFileName); outputTypes["name"].push_back(outputFileName);
478 catch(exception& e) {
479 m->errorOut(e, "GetLineageCommand", "readName");
484 //**********************************************************************************************************************
485 int GetLineageCommand::readGroup(){
487 string thisOutputDir = outputDir;
488 if (outputDir == "") { thisOutputDir += m->hasPath(groupfile); }
489 string outputFileName = thisOutputDir + m->getRootName(m->getSimpleName(groupfile)) + "pick" + m->getExtension(groupfile);
491 m->openOutputFile(outputFileName, out);
495 m->openInputFile(groupfile, in);
498 bool wroteSomething = false;
502 if (m->control_pressed) { in.close(); out.close(); remove(outputFileName.c_str()); return 0; }
505 in >> name; //read from first column
506 in >> group; //read from second column
508 //if this name is in the accnos file
509 if (names.count(name) != 0) {
510 wroteSomething = true;
512 out << name << '\t' << group << endl;
520 if (wroteSomething == false) { m->mothurOut("Your file does not contain any sequence from the .accnos file."); m->mothurOutEndLine(); }
521 outputNames.push_back(outputFileName); outputTypes["group"].push_back(outputFileName);
526 catch(exception& e) {
527 m->errorOut(e, "GetLineageCommand", "readGroup");
531 //**********************************************************************************************************************
532 int GetLineageCommand::readTax(){
534 string thisOutputDir = outputDir;
535 if (outputDir == "") { thisOutputDir += m->hasPath(taxfile); }
536 string outputFileName = thisOutputDir + m->getRootName(m->getSimpleName(taxfile)) + "pick" + m->getExtension(taxfile);
538 m->openOutputFile(outputFileName, out);
541 m->openInputFile(taxfile, in);
544 //bool wroteSomething = false;
546 bool taxonsHasConfidence = false;
547 vector< map<string, int> > searchTaxons;
548 string noConfidenceTaxons = taxons;
549 int hasConPos = taxons.find_first_of('(');
550 if (hasConPos != string::npos) {
551 taxonsHasConfidence = true;
552 searchTaxons = getTaxons(taxons);
553 noConfidenceTaxons = removeConfidences(taxons);
559 if (m->control_pressed) { in.close(); out.close(); remove(outputFileName.c_str()); return 0; }
561 in >> name; //read from first column
562 in >> tax; //read from second column
567 //if the users file contains confidence scores we want to ignore them when searching for the taxons, unless the taxon has them
568 if (!taxonsHasConfidence) {
569 int hasConfidences = tax.find_first_of('(');
570 if (hasConfidences != string::npos) {
571 newtax = removeConfidences(tax);
574 int pos = newtax.find(taxons);
576 if (pos != string::npos) { //this sequence contains the taxon the user wants
578 out << name << '\t' << tax << endl;
581 }else{//if taxons has them and you don't them remove taxons
582 int hasConfidences = tax.find_first_of('(');
583 if (hasConfidences == string::npos) {
585 int pos = newtax.find(noConfidenceTaxons);
587 if (pos != string::npos) { //this sequence contains the taxon the user wants
589 out << name << '\t' << tax << endl;
591 }else { //both have confidences so we want to make sure the users confidences are greater then or equal to the taxons
592 //first remove confidences from both and see if the taxonomy exists
594 string noNewTax = tax;
595 int hasConfidences = tax.find_first_of('(');
596 if (hasConfidences != string::npos) {
597 noNewTax = removeConfidences(tax);
600 int pos = noNewTax.find(noConfidenceTaxons);
602 if (pos != string::npos) { //if yes, then are the confidences okay
605 vector< map<string, int> > usersTaxon = getTaxons(newtax);
607 //the usersTaxon is most likely longer than the searchTaxons, and searchTaxon[0] may relate to userTaxon[4]
608 //we want to "line them up", so we will find the the index where the searchstring starts
610 for (int i = 0; i < usersTaxon.size(); i++) {
612 if (usersTaxon[i].begin()->first == searchTaxons[0].begin()->first) {
615 bool goodspot = true;
616 //is this really the start, or are we dealing with a taxon of the same name?
617 while ((spot < searchTaxons.size()) && ((i+spot) < usersTaxon.size())) {
618 if (usersTaxon[i+spot].begin()->first != searchTaxons[spot].begin()->first) { goodspot = false; break; }
622 if (goodspot) { break; }
626 for (int i = 0; i < searchTaxons.size(); i++) {
628 if ((i+index) < usersTaxon.size()) { //just in case, should never be false
629 if (usersTaxon[i+index].begin()->second < searchTaxons[i].begin()->second) { //is the users cutoff less than the search taxons
639 //passed the test so add you
642 out << name << '\t' << tax << endl;
655 if (names.size() == 0) { m->mothurOut("Your taxonomy file does not contain any sequences from " + taxons + "."); m->mothurOutEndLine(); }
656 outputNames.push_back(outputFileName); outputTypes["taxonomy"].push_back(outputFileName);
661 catch(exception& e) {
662 m->errorOut(e, "GetLineageCommand", "readTax");
666 /**************************************************************************************************/
667 vector< map<string, int> > GetLineageCommand::getTaxons(string tax) {
670 vector< map<string, int> > t;
672 int taxLength = tax.length();
673 for(int i=0;i<taxLength;i++){
676 int openParen = taxon.find_first_of('(');
677 int closeParen = taxon.find_last_of(')');
679 string newtaxon, confidence;
680 if ((openParen != string::npos) && (closeParen != string::npos)) {
681 newtaxon = taxon.substr(0, openParen); //rip off confidence
682 confidence = taxon.substr((openParen+1), (closeParen-openParen-1));
688 convert(confidence, con);
690 map<string, int> temp;
691 temp[newtaxon] = con;
703 catch(exception& e) {
704 m->errorOut(e, "GetLineageCommand", "getTaxons");
708 /**************************************************************************************************/
709 string GetLineageCommand::removeConfidences(string tax) {
713 int taxLength = tax.length();
714 for(int i=0;i<taxLength;i++){
716 taxon = taxon.substr(0, taxon.find_first_of('(')); //rip off confidence
726 catch(exception& e) {
727 m->errorOut(e, "GetLineageCommand", "removeConfidences");
731 //**********************************************************************************************************************
732 //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
733 int GetLineageCommand::readAlign(){
735 string thisOutputDir = outputDir;
736 if (outputDir == "") { thisOutputDir += m->hasPath(alignfile); }
737 string outputFileName = thisOutputDir + m->getRootName(m->getSimpleName(alignfile)) + "pick.align.report";
739 m->openOutputFile(outputFileName, out);
743 m->openInputFile(alignfile, in);
746 bool wroteSomething = false;
748 //read column headers
749 for (int i = 0; i < 16; i++) {
750 if (!in.eof()) { in >> junk; out << junk << '\t'; }
757 if (m->control_pressed) { in.close(); out.close(); remove(outputFileName.c_str()); return 0; }
760 in >> name; //read from first column
762 //if this name is in the accnos file
763 if (names.count(name) != 0) {
764 wroteSomething = true;
769 for (int i = 0; i < 15; i++) {
770 if (!in.eof()) { in >> junk; out << junk << '\t'; }
775 }else {//still read just don't do anything with it
777 for (int i = 0; i < 15; i++) {
778 if (!in.eof()) { in >> junk; }
788 if (wroteSomething == false) { m->mothurOut("Your file does not contain any sequence from the .accnos file."); m->mothurOutEndLine(); }
789 outputNames.push_back(outputFileName); outputTypes["alignreport"].push_back(outputFileName);
794 catch(exception& e) {
795 m->errorOut(e, "GetLineageCommand", "readAlign");
799 //**********************************************************************************************************************