CommandParameter ptaxonomy("taxonomy", "InputTypes", "", "", "none", "none", "none",false,true); parameters.push_back(ptaxonomy);
CommandParameter ptemplate("reference", "InputTypes", "", "", "none", "none", "none",false,true); parameters.push_back(ptemplate);
CommandParameter pfasta("fasta", "InputTypes", "", "", "none", "none", "none",false,true); parameters.push_back(pfasta);
- CommandParameter pname("name", "InputTypes", "", "", "none", "none", "none",false,false); parameters.push_back(pname);
- CommandParameter pgroup("group", "InputTypes", "", "", "none", "none", "none",false,false); parameters.push_back(pgroup);
+ CommandParameter pname("name", "InputTypes", "", "", "NameCount", "none", "none",false,false); parameters.push_back(pname);
+ CommandParameter pcount("count", "InputTypes", "", "", "NameCount-CountGroup", "none", "none",false,false); parameters.push_back(pcount);
+ CommandParameter pgroup("group", "InputTypes", "", "", "CountGroup", "none", "none",false,false); parameters.push_back(pgroup);
+
CommandParameter psearch("search", "Multiple", "kmer-blast-suffix-distance", "kmer", "", "", "",false,false); parameters.push_back(psearch);
CommandParameter pksize("ksize", "Number", "", "8", "", "", "",false,false); parameters.push_back(pksize);
CommandParameter pmethod("method", "Multiple", "bayesian-knn", "bayesian", "", "", "",false,false); parameters.push_back(pmethod);
CommandParameter pprobs("probs", "Boolean", "", "T", "", "", "",false,false); parameters.push_back(pprobs);
CommandParameter piters("iters", "Number", "", "100", "", "", "",false,true); parameters.push_back(piters);
CommandParameter psave("save", "Boolean", "", "F", "", "", "",false,false); parameters.push_back(psave);
+ CommandParameter pshortcuts("shortcuts", "Boolean", "", "T", "", "", "",false,false); parameters.push_back(pshortcuts);
CommandParameter pnumwanted("numwanted", "Number", "", "10", "", "", "",false,true); parameters.push_back(pnumwanted);
CommandParameter pinputdir("inputdir", "String", "", "", "", "", "",false,false); parameters.push_back(pinputdir);
CommandParameter poutputdir("outputdir", "String", "", "", "", "", "",false,false); parameters.push_back(poutputdir);
try {
string helpString = "";
helpString += "The classify.seqs command reads a fasta file containing sequences and creates a .taxonomy file and a .tax.summary file.\n";
- helpString += "The classify.seqs command parameters are reference, fasta, name, search, ksize, method, taxonomy, processors, match, mismatch, gapopen, gapextend, numwanted and probs.\n";
+ helpString += "The classify.seqs command parameters are reference, fasta, name, group, count, search, ksize, method, taxonomy, processors, match, mismatch, gapopen, gapextend, numwanted and probs.\n";
helpString += "The reference, fasta and taxonomy parameters are required. You may enter multiple fasta files by separating their names with dashes. ie. fasta=abrecovery.fasta-amzon.fasta \n";
helpString += "The search parameter allows you to specify the method to find most similar template. Your options are: suffix, kmer, blast and distance. The default is kmer.\n";
helpString += "The name parameter allows you add a names file with your fasta file, if you enter multiple fasta files, you must enter matching names files for them.\n";
helpString += "The group parameter allows you add a group file so you can have the summary totals broken up by group.\n";
+ helpString += "The count parameter allows you add a count file so you can have the summary totals broken up by group.\n";
helpString += "The method parameter allows you to specify classification method to use. Your options are: bayesian and knn. The default is bayesian.\n";
helpString += "The ksize parameter allows you to specify the kmer size for finding most similar template to candidate. The default is 8.\n";
helpString += "The processors parameter allows you to specify the number of processors to use. The default is 1.\n";
}
}
//**********************************************************************************************************************
+string ClassifySeqsCommand::getOutputFileNameTag(string type, string inputName=""){
+ try {
+ string outputFileName = "";
+ map<string, vector<string> >::iterator it;
+
+ //is this a type this command creates
+ it = outputTypes.find(type);
+ if (it == outputTypes.end()) { m->mothurOut("[ERROR]: this command doesn't create a " + type + " output file.\n"); }
+ else {
+ if (type == "taxonomy") { outputFileName = "taxonomy"; }
+ else if (type == "accnos") { outputFileName = "flip.accnos"; }
+ else if (type == "taxsummary") { outputFileName = "tax.summary"; }
+ else if (type == "matchdist") { outputFileName = "match.dist"; }
+ else { m->mothurOut("[ERROR]: No definition for type " + type + " output file tag.\n"); m->control_pressed = true; }
+ }
+ return outputFileName;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "ClassifySeqsCommand", "getOutputFileNameTag");
+ exit(1);
+ }
+}
+//**********************************************************************************************************************
ClassifySeqsCommand::ClassifySeqsCommand(){
try {
abort = true; calledHelp = true;
ClassifySeqsCommand::ClassifySeqsCommand(string option) {
try {
abort = false; calledHelp = false;
- rdb = ReferenceDB::getInstance();
+ rdb = ReferenceDB::getInstance(); hasName = false; hasCount=false;
//allow user to run help
if(option == "help") { help(); abort = true; calledHelp = true; }
//if the user has not given a path then, add inputdir. else leave path alone.
if (path == "") { parameters["group"] = inputDir + it->second; }
}
+
+ it = parameters.find("count");
+ //user has given a template file
+ if(it != parameters.end()){
+ path = m->hasPath(it->second);
+ //if the user has not given a path then, add inputdir. else leave path alone.
+ if (path == "") { parameters["count"] = inputDir + it->second; }
+ }
}
fastaFileName = validParameter.validFile(parameters, "fasta", false);
}
}
}
-
+
+ if (namefileNames.size() != 0) { hasName = true; }
+
if (namefile != "") {
if (namefileNames.size() != fastaFileNames.size()) { abort = true; m->mothurOut("If you provide a name file, you must have one for each fasta file."); m->mothurOutEndLine(); }
}
+ //check for required parameters
+ countfile = validParameter.validFile(parameters, "count", false);
+ if (countfile == "not found") {
+ countfile = "";
+ }else {
+ m->splitAtDash(countfile, countfileNames);
+
+ //go through files and make sure they are good, if not, then disregard them
+ for (int i = 0; i < countfileNames.size(); i++) {
+
+ bool ignore = false;
+ if (countfileNames[i] == "current") {
+ countfileNames[i] = m->getCountTableFile();
+ if (countfileNames[i] != "") { m->mothurOut("Using " + countfileNames[i] + " as input file for the count parameter where you had given current."); m->mothurOutEndLine(); }
+ else {
+ m->mothurOut("You have no current count file, ignoring current."); m->mothurOutEndLine(); ignore=true;
+ //erase from file list
+ countfileNames.erase(countfileNames.begin()+i);
+ i--;
+ }
+ }
+
+ if (!ignore) {
+
+ if (inputDir != "") {
+ string path = m->hasPath(countfileNames[i]);
+ //if the user has not given a path then, add inputdir. else leave path alone.
+ if (path == "") { countfileNames[i] = inputDir + countfileNames[i]; }
+ }
+
+ int ableToOpen;
+ ifstream in;
+
+ ableToOpen = m->openInputFile(countfileNames[i], in, "noerror");
+
+ //if you can't open it, try default location
+ if (ableToOpen == 1) {
+ if (m->getDefaultPath() != "") { //default path is set
+ string tryPath = m->getDefaultPath() + m->getSimpleName(countfileNames[i]);
+ m->mothurOut("Unable to open " + countfileNames[i] + ". Trying default " + tryPath); m->mothurOutEndLine();
+ ifstream in2;
+ ableToOpen = m->openInputFile(tryPath, in2, "noerror");
+ in2.close();
+ countfileNames[i] = tryPath;
+ }
+ }
+
+ if (ableToOpen == 1) {
+ if (m->getOutputDir() != "") { //default path is set
+ string tryPath = m->getOutputDir() + m->getSimpleName(countfileNames[i]);
+ m->mothurOut("Unable to open " + countfileNames[i] + ". Trying output directory " + tryPath); m->mothurOutEndLine();
+ ifstream in2;
+ ableToOpen = m->openInputFile(tryPath, in2, "noerror");
+ in2.close();
+ countfileNames[i] = tryPath;
+ }
+ }
+
+ in.close();
+
+ if (ableToOpen == 1) {
+ m->mothurOut("Unable to open " + countfileNames[i] + ". It will be disregarded."); m->mothurOutEndLine();
+ //erase from file list
+ countfileNames.erase(countfileNames.begin()+i);
+ i--;
+ }else {
+ m->setCountTableFile(countfileNames[i]);
+ }
+ }
+ }
+ }
+
+ if (countfileNames.size() != 0) { hasCount = true; if (countfileNames.size() != fastaFileNames.size()) {m->mothurOut("If you provide a count file, you must have one for each fasta file."); m->mothurOutEndLine(); } }
+
+ //make sure there is at least one valid file left
+ if (hasName && hasCount) { m->mothurOut("[ERROR]: You must enter ONLY ONE of the following: count or name."); m->mothurOutEndLine(); abort = true; }
+
groupfile = validParameter.validFile(parameters, "group", false);
if (groupfile == "not found") { groupfile = ""; }
else {
if (groupfile != "") {
if (groupfileNames.size() != fastaFileNames.size()) { abort = true; m->mothurOut("If you provide a group file, you must have one for each fasta file."); m->mothurOutEndLine(); }
+ if (hasCount) { m->mothurOut("[ERROR]: You must enter ONLY ONE of the following: count or group."); m->mothurOutEndLine(); abort = true; }
}else {
for (int i = 0; i < fastaFileNames.size(); i++) { groupfileNames.push_back(""); }
}
temp = validParameter.validFile(parameters, "probs", false); if (temp == "not found"){ temp = "true"; }
probs = m->isTrue(temp);
+
+ temp = validParameter.validFile(parameters, "shortcuts", false); if (temp == "not found"){ temp = "true"; }
+ writeShortcuts = m->isTrue(temp);
//temp = validParameter.validFile(parameters, "flip", false); if (temp == "not found"){ temp = "T"; }
//flip = m->isTrue(temp);
search = "kmer";
}
- if (namefileNames.size() == 0){
- vector<string> files; files.push_back(fastaFileNames[fastaFileNames.size()-1]);
- parser.getNameFile(files);
- }
-
- }
-
+ if (!abort) {
+ if (!hasCount) {
+ if (namefileNames.size() == 0){
+ if (fastaFileNames.size() != 0) {
+ vector<string> files; files.push_back(fastaFileNames[fastaFileNames.size()-1]);
+ parser.getNameFile(files);
+ }
+ }
+ }
+ }
+ }
}
catch(exception& e) {
m->errorOut(e, "ClassifySeqsCommand", "ClassifySeqsCommand");
int ClassifySeqsCommand::execute(){
try {
if (abort == true) { if (calledHelp) { return 0; } return 2; }
-
- if(method == "bayesian"){ classify = new Bayesian(taxonomyFileName, templateFileName, search, kmerSize, cutoff, iters, rand(), flip); }
+
+ if(method == "bayesian"){ classify = new Bayesian(taxonomyFileName, templateFileName, search, kmerSize, cutoff, iters, rand(), flip, writeShortcuts); }
else if(method == "knn"){ classify = new Knn(taxonomyFileName, templateFileName, search, kmerSize, gapOpen, gapExtend, match, misMatch, numWanted, rand()); }
else {
m->mothurOut(search + " is not a valid method option. I will run the command using bayesian.");
m->mothurOutEndLine();
- classify = new Bayesian(taxonomyFileName, templateFileName, search, kmerSize, cutoff, iters, rand(), flip);
+ classify = new Bayesian(taxonomyFileName, templateFileName, search, kmerSize, cutoff, iters, rand(), flip, writeShortcuts);
}
if (m->control_pressed) { delete classify; return 0; }
m->mothurOut("Classifying sequences from " + fastaFileNames[s] + " ..." ); m->mothurOutEndLine();
- string baseTName = taxonomyFileName;
+ string baseTName = m->getSimpleName(taxonomyFileName);
if (taxonomyFileName == "saved") {baseTName = rdb->getSavedTaxonomy(); }
- string RippedTaxName = m->getRootName(m->getSimpleName(baseTName));
- RippedTaxName = m->getExtension(RippedTaxName.substr(0, RippedTaxName.length()-1));
- if (RippedTaxName[0] == '.') { RippedTaxName = RippedTaxName.substr(1, RippedTaxName.length()); }
- RippedTaxName += ".";
-
+ //set rippedTaxName to
+ string RippedTaxName = "";
+ bool foundDot = false;
+ for (int i = baseTName.length()-1; i >= 0; i--) {
+ if (foundDot && (baseTName[i] != '.')) { RippedTaxName = baseTName[i] + RippedTaxName; }
+ else if (foundDot && (baseTName[i] == '.')) { break; }
+ else if (!foundDot && (baseTName[i] == '.')) { foundDot = true; }
+ }
+ if (RippedTaxName != "") { RippedTaxName += "."; }
+
if (outputDir == "") { outputDir += m->hasPath(fastaFileNames[s]); }
- string newTaxonomyFile = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + RippedTaxName + "taxonomy";
- string newaccnosFile = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + RippedTaxName + "flip.accnos";
+ string newTaxonomyFile = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + RippedTaxName + getOutputFileNameTag("taxonomy");
+ string newaccnosFile = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + RippedTaxName + getOutputFileNameTag("accnos");
string tempTaxonomyFile = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + "taxonomy.temp";
- string taxSummary = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + RippedTaxName + "tax.summary";
+ string taxSummary = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + RippedTaxName + getOutputFileNameTag("taxsummary");
if ((method == "knn") && (search == "distance")) {
- string DistName = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + "match.dist";
+ string DistName = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + getOutputFileNameTag("matchdist");
classify->setDistName(DistName); outputNames.push_back(DistName); outputTypes["matchdist"].push_back(DistName);
}
if(namefile != "") {
m->mothurOut("Reading " + namefileNames[s] + "..."); cout.flush();
-
nameMap.clear(); //remove old names
-
- ifstream inNames;
- m->openInputFile(namefileNames[s], inNames);
-
- string firstCol, secondCol;
- while(!inNames.eof()) {
- inNames >> firstCol >> secondCol; m->gobble(inNames);
-
- vector<string> temp;
- m->splitAtComma(secondCol, temp);
-
- nameMap[firstCol] = temp;
- }
- inNames.close();
-
+ m->readNames(namefileNames[s], nameMap);
m->mothurOut(" Done."); m->mothurOutEndLine();
}
#endif
- string group = "";
- if (groupfile != "") { group = groupfileNames[s]; }
-
- PhyloSummary taxaSum(baseTName, group);
-
- if (m->control_pressed) { outputTypes.clear(); for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } delete classify; return 0; }
-
- if (namefile == "") { taxaSum.summarize(tempTaxonomyFile); }
- else {
- ifstream in;
- m->openInputFile(tempTaxonomyFile, in);
-
- //read in users taxonomy file and add sequences to tree
- string name, taxon;
-
- while(!in.eof()){
- in >> name >> taxon; m->gobble(in);
-
- itNames = nameMap.find(name);
-
- if (itNames == nameMap.end()) {
- m->mothurOut(name + " is not in your name file please correct."); m->mothurOutEndLine(); exit(1);
- }else{
- for (int i = 0; i < itNames->second.size(); i++) {
- taxaSum.addSeqToTree(itNames->second[i], taxon); //add it as many times as there are identical seqs
- }
- itNames->second.clear();
- nameMap.erase(itNames->first);
- }
- }
- in.close();
- }
+ string group = "";
+ GroupMap* groupMap = NULL;
+ CountTable* ct = NULL;
+ PhyloSummary* taxaSum;
+ if (hasCount) {
+ ct = new CountTable();
+ ct->readTable(countfileNames[s]);
+ taxaSum = new PhyloSummary(baseTName, ct);
+ taxaSum->summarize(tempTaxonomyFile);
+ }else {
+ if (groupfile != "") { group = groupfileNames[s]; groupMap = new GroupMap(group); groupMap->readMap(); }
+
+ taxaSum = new PhyloSummary(baseTName, groupMap);
+
+ if (m->control_pressed) { outputTypes.clear(); if (ct != NULL) { delete ct; } if (groupMap != NULL) { delete groupMap; } delete taxaSum; for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } delete classify; return 0; }
+
+ if (namefile == "") { taxaSum->summarize(tempTaxonomyFile); }
+ else {
+ ifstream in;
+ m->openInputFile(tempTaxonomyFile, in);
+
+ //read in users taxonomy file and add sequences to tree
+ string name, taxon;
+
+ while(!in.eof()){
+ if (m->control_pressed) { outputTypes.clear(); if (ct != NULL) { delete ct; } if (groupMap != NULL) { delete groupMap; } delete taxaSum; for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } delete classify; return 0; }
+
+ in >> name >> taxon; m->gobble(in);
+
+ itNames = nameMap.find(name);
+
+ if (itNames == nameMap.end()) {
+ m->mothurOut(name + " is not in your name file please correct."); m->mothurOutEndLine(); exit(1);
+ }else{
+ for (int i = 0; i < itNames->second.size(); i++) {
+ taxaSum->addSeqToTree(itNames->second[i], taxon); //add it as many times as there are identical seqs
+ }
+ itNames->second.clear();
+ nameMap.erase(itNames->first);
+ }
+ }
+ in.close();
+ }
+ }
m->mothurRemove(tempTaxonomyFile);
- if (m->control_pressed) { outputTypes.clear(); for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } delete classify; return 0; }
+ if (m->control_pressed) { outputTypes.clear(); if (ct != NULL) { delete ct; } if (groupMap != NULL) { delete groupMap; } for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } delete classify; return 0; }
//print summary file
ofstream outTaxTree;
m->openOutputFile(taxSummary, outTaxTree);
- taxaSum.print(outTaxTree);
+ taxaSum->print(outTaxTree);
outTaxTree.close();
//output taxonomy with the unclassified bins added
m->openOutputFile(unclass, outTax);
//get maxLevel from phylotree so you know how many 'unclassified's to add
- int maxLevel = taxaSum.getMaxLevel();
+ int maxLevel = taxaSum->getMaxLevel();
//read taxfile - this reading and rewriting is done to preserve the confidence scores.
string name, taxon;
while (!inTax.eof()) {
- if (m->control_pressed) { outputTypes.clear(); for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } m->mothurRemove(unclass); delete classify; return 0; }
+ if (m->control_pressed) { outputTypes.clear(); if (ct != NULL) { delete ct; } if (groupMap != NULL) { delete groupMap; } delete taxaSum; for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } m->mothurRemove(unclass); delete classify; return 0; }
inTax >> name >> taxon; m->gobble(inTax);
inTax.close();
outTax.close();
+ if (ct != NULL) { delete ct; }
+ if (groupMap != NULL) { delete groupMap; } delete taxaSum;
m->mothurRemove(newTaxonomyFile);
rename(unclass.c_str(), newTaxonomyFile.c_str());
string extension = "";
if (i != 0) { extension = toString(i) + ".temp"; processIDS.push_back(i); }
- classifyData* tempclass = new classifyData((accnos + extension), probs, method, templateFileName, taxonomyFileName, (taxFileName + extension), (tempTaxFile + extension), filename, search, kmerSize, iters, numWanted, m, lines[i]->start, lines[i]->end, match, misMatch, gapOpen, gapExtend, cutoff, i, flip);
+ classifyData* tempclass = new classifyData((accnos + extension), probs, method, templateFileName, taxonomyFileName, (taxFileName + extension), (tempTaxFile + extension), filename, search, kmerSize, iters, numWanted, m, lines[i]->start, lines[i]->end, match, misMatch, gapOpen, gapExtend, cutoff, i, flip, writeShortcuts);
pDataArray.push_back(tempclass);
//MySeqSumThreadFunction is in header. It must be global or static to work with the threads.
else { m->mothurRemove(accnos); } //remove so other files can be renamed to it
for(int i=0;i<processIDS.size();i++){
- appendTaxFiles((taxFileName + toString(processIDS[i]) + ".temp"), taxFileName);
- appendTaxFiles((tempTaxFile + toString(processIDS[i]) + ".temp"), tempTaxFile);
+ m->appendFiles((taxFileName + toString(processIDS[i]) + ".temp"), taxFileName);
+ m->appendFiles((tempTaxFile + toString(processIDS[i]) + ".temp"), tempTaxFile);
if (!(m->isBlank(accnos + toString(processIDS[i]) + ".temp"))) {
nonBlankAccnosFiles.push_back(accnos + toString(processIDS[i]) + ".temp");
}else { m->mothurRemove((accnos + toString(processIDS[i]) + ".temp")); }
rename(nonBlankAccnosFiles[0].c_str(), accnos.c_str());
for (int h=1; h < nonBlankAccnosFiles.size(); h++) {
- appendTaxFiles(nonBlankAccnosFiles[h], accnos);
+ m->appendFiles(nonBlankAccnosFiles[h], accnos);
m->mothurRemove(nonBlankAccnosFiles[h]);
}
}else { //recreate the accnosfile if needed
exit(1);
}
}
-/**************************************************************************************************/
-
-void ClassifySeqsCommand::appendTaxFiles(string temp, string filename) {
- try{
-
- ofstream output;
- ifstream input;
- m->openOutputFileAppend(filename, output);
- m->openInputFile(temp, input);
-
- while(char c = input.get()){
- if(input.eof()) { break; }
- else { output << c; }
- }
-
- input.close();
- output.close();
- }
- catch(exception& e) {
- m->errorOut(e, "ClassifySeqsCommand", "appendTaxFiles");
- exit(1);
- }
-}
-
//**********************************************************************************************************************
int ClassifySeqsCommand::driver(linePair* filePos, string taxFName, string tempTFName, string accnos, string filename){
projects / mothur.git / blobdiff