X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=removelineagecommand.cpp;h=021b56cd02f36aaa88edac90a8773ed1090bc5a0;hb=ca9ac1d80c62f57270b0dcd49410ebe08a8aecd6;hp=629416a5f01476442b3dd3e8db9cd3fadaca4165;hpb=df41d90a9611eab75d0516e2654ed8580f8df24c;p=mothur.git diff --git a/removelineagecommand.cpp b/removelineagecommand.cpp index 629416a..021b56c 100644 --- a/removelineagecommand.cpp +++ b/removelineagecommand.cpp @@ -12,21 +12,58 @@ #include "listvector.hpp" //********************************************************************************************************************** -vector RemoveLineageCommand::getValidParameters(){ +vector RemoveLineageCommand::setParameters(){ try { - string Array[] = {"fasta","name", "group", "alignreport", "taxon", "dups", "list","taxonomy","outputdir","inputdir"}; - vector myArray (Array, Array+(sizeof(Array)/sizeof(string))); + CommandParameter pfasta("fasta", "InputTypes", "", "", "none", "FNGLT", "none",false,false); parameters.push_back(pfasta); + CommandParameter pname("name", "InputTypes", "", "", "none", "FNGLT", "none",false,false); parameters.push_back(pname); + CommandParameter pgroup("group", "InputTypes", "", "", "none", "FNGLT", "none",false,false); parameters.push_back(pgroup); + CommandParameter plist("list", "InputTypes", "", "", "none", "FNGLT", "none",false,false); parameters.push_back(plist); + CommandParameter ptaxonomy("taxonomy", "InputTypes", "", "", "none", "FNGLT", "none",false,true); parameters.push_back(ptaxonomy); + CommandParameter palignreport("alignreport", "InputTypes", "", "", "none", "FNGLT", "none",false,false); parameters.push_back(palignreport); + CommandParameter ptaxon("taxon", "String", "", "", "", "", "",false,true); parameters.push_back(ptaxon); + CommandParameter pdups("dups", "Boolean", "", "T", "", "", "",false,false); parameters.push_back(pdups); + CommandParameter pinputdir("inputdir", "String", "", "", "", "", "",false,false); parameters.push_back(pinputdir); + CommandParameter poutputdir("outputdir", "String", "", "", "", "", "",false,false); parameters.push_back(poutputdir); + + vector myArray; + for (int i = 0; i < parameters.size(); i++) { myArray.push_back(parameters[i].name); } return myArray; } catch(exception& e) { - m->errorOut(e, "RemoveLineageCommand", "getValidParameters"); + m->errorOut(e, "RemoveLineageCommand", "setParameters"); exit(1); } } +//********************************************************************************************************************** +string RemoveLineageCommand::getHelpString(){ + try { + string helpString = ""; + helpString += "The remove.lineage command reads a taxonomy file and any of the following file types: fasta, name, group, list or alignreport file.\n"; + helpString += "It outputs a file containing only the sequences from the taxonomy file that are not from the taxon you requested to be removed.\n"; + helpString += "The remove.lineage command parameters are taxon, fasta, name, group, list, taxonomy, alignreport and dups. You must provide taxonomy unless you have a valid current taxonomy file.\n"; + helpString += "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"; + helpString += "The taxon parameter allows you to select the taxons you would like to remove, and is required.\n"; + helpString += "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"; + helpString += "If they belong to the taxonomy and have confidences above those you provide the sequence will not be removed.\n"; + helpString += "The remove.lineage command should be in the following format: remove.lineage(taxonomy=yourTaxonomyFile, taxon=yourTaxons).\n"; + helpString += "Example remove.lineage(taxonomy=amazon.silva.taxonomy, taxon=Bacteria;Firmicutes;Bacilli;Lactobacillales;).\n"; + helpString += "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"; + helpString += "Example remove.lineage(taxonomy=amazon.silva.taxonomy, taxon='Bacteria;Firmicutes;Bacilli;Lactobacillales;').\n"; + helpString += "Note: No spaces between parameter labels (i.e. fasta), '=' and parameters (i.e.yourFasta).\n\n"; + return helpString; + } + catch(exception& e) { + m->errorOut(e, "RemoveLineageCommand", "getHelpString"); + exit(1); + } +} + + //********************************************************************************************************************** RemoveLineageCommand::RemoveLineageCommand(){ try { - //initialize outputTypes + abort = true; calledHelp = true; + setParameters(); vector tempOutNames; outputTypes["fasta"] = tempOutNames; outputTypes["taxonomy"] = tempOutNames; @@ -41,35 +78,12 @@ RemoveLineageCommand::RemoveLineageCommand(){ } } //********************************************************************************************************************** -vector RemoveLineageCommand::getRequiredParameters(){ - try { - string Array[] = {"taxonomy"}; - vector myArray (Array, Array+(sizeof(Array)/sizeof(string))); - return myArray; - } - catch(exception& e) { - m->errorOut(e, "RemoveLineageCommand", "getRequiredParameters"); - exit(1); - } -} -//********************************************************************************************************************** -vector RemoveLineageCommand::getRequiredFiles(){ - try { - vector myArray; - return myArray; - } - catch(exception& e) { - m->errorOut(e, "RemoveLineageCommand", "getRequiredFiles"); - exit(1); - } -} -//********************************************************************************************************************** RemoveLineageCommand::RemoveLineageCommand(string option) { try { - abort = false; + abort = false; calledHelp = false; //allow user to run help - if(option == "help") { help(); abort = true; } + if(option == "help") { help(); abort = true; calledHelp = true; } else { //valid paramters for this command @@ -177,10 +191,18 @@ RemoveLineageCommand::RemoveLineageCommand(string option) { taxfile = validParameter.validFile(parameters, "taxonomy", true); if (taxfile == "not open") { abort = true; } - else if (taxfile == "not found") { taxfile = ""; m->mothurOut("The taxonomy parameter is required for the get.lineage command."); m->mothurOutEndLine(); abort = true; } + else if (taxfile == "not found") { + taxfile = m->getTaxonomyFile(); + if (taxfile != "") { m->mothurOut("Using " + taxfile + " as input file for the taxonomy parameter."); m->mothurOutEndLine(); } + else { m->mothurOut("You have no current taxonomy file and the taxonomy parameter is required."); m->mothurOutEndLine(); abort = true; } + } string usedDups = "true"; - string temp = validParameter.validFile(parameters, "dups", false); if (temp == "not found") { temp = "false"; usedDups = ""; } + string temp = validParameter.validFile(parameters, "dups", false); + if (temp == "not found") { + if (namefile != "") { temp = "true"; } + else { temp = "false"; usedDups = ""; } + } dups = m->isTrue(temp); taxons = validParameter.validFile(parameters, "taxon", false); @@ -206,31 +228,10 @@ RemoveLineageCommand::RemoveLineageCommand(string option) { } //********************************************************************************************************************** -void RemoveLineageCommand::help(){ - try { - 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"); - 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"); - m->mothurOut("The remove.lineage command parameters are taxon, fasta, name, group, list, taxonomy, alignreport and dups. You must provide taxonomy and taxon.\n"); - 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"); - m->mothurOut("The taxon parameter allows you to select the taxons you would like to remove.\n"); - m->mothurOut("The remove.lineage command should be in the following format: remove.lineage(taxonomy=yourTaxonomyFile, taxon=yourTaxons).\n"); - m->mothurOut("Example remove.lineage(taxonomy=amazon.silva.taxonomy, taxon=Bacteria;Firmicutes;Bacilli;Lactobacillales;).\n"); - 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"); - m->mothurOut("Example remove.lineage(taxonomy=amazon.silva.taxonomy, taxon='Bacteria;Firmicutes;Bacilli;Lactobacillales;').\n"); - m->mothurOut("Note: No spaces between parameter labels (i.e. fasta), '=' and parameters (i.e.yourFasta).\n\n"); - } - catch(exception& e) { - m->errorOut(e, "RemoveLineageCommand", "help"); - exit(1); - } -} - -//********************************************************************************************************************** - int RemoveLineageCommand::execute(){ try { - if (abort == true) { return 0; } + if (abort == true) { if (calledHelp) { return 0; } return 2; } if (m->control_pressed) { return 0; } @@ -250,6 +251,33 @@ int RemoveLineageCommand::execute(){ m->mothurOut("Output File Names: "); m->mothurOutEndLine(); for (int i = 0; i < outputNames.size(); i++) { m->mothurOut(outputNames[i]); m->mothurOutEndLine(); } m->mothurOutEndLine(); + + //set fasta file as new current fastafile + string current = ""; + itTypes = outputTypes.find("fasta"); + if (itTypes != outputTypes.end()) { + if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setFastaFile(current); } + } + + itTypes = outputTypes.find("name"); + if (itTypes != outputTypes.end()) { + if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setNameFile(current); } + } + + itTypes = outputTypes.find("group"); + if (itTypes != outputTypes.end()) { + if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setGroupFile(current); } + } + + itTypes = outputTypes.find("list"); + if (itTypes != outputTypes.end()) { + if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setListFile(current); } + } + + itTypes = outputTypes.find("taxonomy"); + if (itTypes != outputTypes.end()) { + if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setTaxonomyFile(current); } + } } return 0; @@ -401,16 +429,8 @@ int RemoveLineageCommand::readName(){ in >> secondCol; vector parsedNames; - //parse second column saving each name - while (secondCol.find_first_of(',') != -1) { - name = secondCol.substr(0,secondCol.find_first_of(',')); - secondCol = secondCol.substr(secondCol.find_first_of(',')+1, secondCol.length()); - parsedNames.push_back(name); - } + m->splitAtComma(secondCol, parsedNames); - //get name after last , - parsedNames.push_back(secondCol); - vector validSecond; validSecond.clear(); for (int i = 0; i < parsedNames.size(); i++) { if (names.count(parsedNames[i]) == 0) { @@ -522,6 +542,17 @@ int RemoveLineageCommand::readTax(){ bool wroteSomething = false; + bool taxonsHasConfidence = false; + vector< map > searchTaxons; + string noConfidenceTaxons = taxons; + int hasConPos = taxons.find_first_of('('); + if (hasConPos != string::npos) { + taxonsHasConfidence = true; + searchTaxons = getTaxons(taxons); + noConfidenceTaxons = removeConfidences(taxons); + } + + while(!in.eof()){ if (m->control_pressed) { in.close(); out.close(); remove(outputFileName.c_str()); return 0; } @@ -531,20 +562,99 @@ int RemoveLineageCommand::readTax(){ string newtax = tax; - //if the users file contains confidence scores we want to ignore them when searching for the taxons - int hasConfidences = tax.find_first_of('('); - if (hasConfidences != string::npos) { - newtax = removeConfidences(tax); + //if the users file contains confidence scores we want to ignore them when searching for the taxons, unless the taxon has them + if (!taxonsHasConfidence) { + + int hasConfidences = tax.find_first_of('('); + if (hasConfidences != string::npos) { + newtax = removeConfidences(tax); + } + + int pos = newtax.find(taxons); + + if (pos == string::npos) { + wroteSomething = true; + out << name << '\t' << tax << endl; + }else{ //this sequence contains the taxon the user wants to remove + names.insert(name); + } + + }else{//if taxons has them and you don't them remove taxons + int hasConfidences = tax.find_first_of('('); + if (hasConfidences == string::npos) { + + int pos = newtax.find(noConfidenceTaxons); + + if (pos == string::npos) { + wroteSomething = true; + out << name << '\t' << tax << endl; + }else{ //this sequence contains the taxon the user wants to remove + names.insert(name); + } + }else { //both have confidences so we want to make sure the users confidences are greater then or equal to the taxons + //first remove confidences from both and see if the taxonomy exists + + string noNewTax = tax; + int hasConfidences = tax.find_first_of('('); + if (hasConfidences != string::npos) { + noNewTax = removeConfidences(tax); + } + + int pos = noNewTax.find(noConfidenceTaxons); + + if (pos != string::npos) { //if yes, then are the confidences okay + + bool remove = false; + vector< map > usersTaxon = getTaxons(newtax); + + //the usersTaxon is most likely longer than the searchTaxons, and searchTaxon[0] may relate to userTaxon[4] + //we want to "line them up", so we will find the the index where the searchstring starts + int index = 0; + for (int i = 0; i < usersTaxon.size(); i++) { + + if (usersTaxon[i].begin()->first == searchTaxons[0].begin()->first) { + index = i; + int spot = 0; + bool goodspot = true; + //is this really the start, or are we dealing with a taxon of the same name? + while ((spot < searchTaxons.size()) && ((i+spot) < usersTaxon.size())) { + if (usersTaxon[i+spot].begin()->first != searchTaxons[spot].begin()->first) { goodspot = false; break; } + else { spot++; } + } + + if (goodspot) { break; } + } + } + + for (int i = 0; i < searchTaxons.size(); i++) { + + if ((i+index) < usersTaxon.size()) { //just in case, should never be false + if (usersTaxon[i+index].begin()->second < searchTaxons[i].begin()->second) { //is the users cutoff less than the search taxons + remove = true; + break; + } + }else { + remove = true; + break; + } + } + + //passed the test so remove you + if (remove) { + names.insert(name); + }else { + wroteSomething = true; + out << name << '\t' << tax << endl; + } + }else { + wroteSomething = true; + out << name << '\t' << tax << endl; + } + } } - int pos = newtax.find(taxons); - if (pos == string::npos) { - wroteSomething = true; - out << name << '\t' << tax << endl; - }else{ //this sequence contains the taxon the user wants to remove - names.insert(name); - } + m->gobble(in); } @@ -563,6 +673,48 @@ int RemoveLineageCommand::readTax(){ } } /**************************************************************************************************/ +vector< map > RemoveLineageCommand::getTaxons(string tax) { + try { + + vector< map > t; + string taxon = ""; + int taxLength = tax.length(); + for(int i=0;i temp; + temp[newtaxon] = con; + t.push_back(temp); + + taxon = ""; + } + else{ + taxon += tax[i]; + } + } + + return t; + } + catch(exception& e) { + m->errorOut(e, "RemoveLineageCommand", "getTaxons"); + exit(1); + } +} +/**************************************************************************************************/ string RemoveLineageCommand::removeConfidences(string tax) { try {