//**********************************************************************************************************************
-ClassifySeqsCommand::ClassifySeqsCommand(string option){
+ClassifySeqsCommand::ClassifySeqsCommand(string option) {
try {
abort = false;
else {
//valid paramters for this command
- string AlignArray[] = {"template","fasta","name","search","ksize","method","processors","taxonomy","match","mismatch","gapopen","gapextend","numwanted","cutoff","probs","iters"};
+ string AlignArray[] = {"template","fasta","name","search","ksize","method","processors","taxonomy","match","mismatch","gapopen","gapextend","numwanted","cutoff","probs","iters", "outputdir","inputdir"};
vector<string> myArray (AlignArray, AlignArray+(sizeof(AlignArray)/sizeof(string)));
OptionParser parser(option);
map<string, string> parameters = parser.getParameters();
ValidParameters validParameter;
+ map<string, string>::iterator it;
//check to make sure all parameters are valid for command
- for (map<string, string>::iterator it = parameters.begin(); it != parameters.end(); it++) {
+ for (it = parameters.begin(); it != parameters.end(); it++) {
if (validParameter.isValidParameter(it->first, myArray, it->second) != true) { abort = true; }
}
+ //if the user changes the output directory command factory will send this info to us in the output parameter
+ outputDir = validParameter.validFile(parameters, "outputdir", false); if (outputDir == "not found"){ outputDir = ""; }
+
+ //if the user changes the input directory command factory will send this info to us in the output parameter
+ string inputDir = validParameter.validFile(parameters, "inputdir", false);
+ if (inputDir == "not found"){ inputDir = ""; }
+ else {
+ string path;
+ it = parameters.find("template");
+ //user has given a template file
+ if(it != parameters.end()){
+ path = hasPath(it->second);
+ //if the user has not given a path then, add inputdir. else leave path alone.
+ if (path == "") { parameters["template"] = inputDir + it->second; }
+ }
+
+ it = parameters.find("taxonomy");
+ //user has given a template file
+ if(it != parameters.end()){
+ path = hasPath(it->second);
+ //if the user has not given a path then, add inputdir. else leave path alone.
+ if (path == "") { parameters["taxonomy"] = inputDir + it->second; }
+ }
+ }
+
//check for required parameters
templateFileName = validParameter.validFile(parameters, "template", true);
if (templateFileName == "not found") {
- mothurOut("template is a required parameter for the classify.seqs command.");
- mothurOutEndLine();
+ m->mothurOut("template is a required parameter for the classify.seqs command.");
+ m->mothurOutEndLine();
abort = true;
}
else if (templateFileName == "not open") { abort = true; }
fastaFileName = validParameter.validFile(parameters, "fasta", false);
- if (fastaFileName == "not found") { mothurOut("fasta is a required parameter for the classify.seqs command."); mothurOutEndLine(); abort = true; }
+ if (fastaFileName == "not found") { m->mothurOut("fasta is a required parameter for the classify.seqs command."); m->mothurOutEndLine(); abort = true; }
else {
splitAtDash(fastaFileName, fastaFileNames);
//go through files and make sure they are good, if not, then disregard them
for (int i = 0; i < fastaFileNames.size(); i++) {
+ if (inputDir != "") {
+ string path = hasPath(fastaFileNames[i]);
+ //if the user has not given a path then, add inputdir. else leave path alone.
+ if (path == "") { fastaFileNames[i] = inputDir + fastaFileNames[i]; }
+ }
+
int ableToOpen;
ifstream in;
ableToOpen = openInputFile(fastaFileNames[i], in);
if (ableToOpen == 1) {
- mothurOut(fastaFileNames[i] + " will be disregarded."); mothurOutEndLine();
+ m->mothurOut(fastaFileNames[i] + " will be disregarded."); m->mothurOutEndLine();
//erase from file list
fastaFileNames.erase(fastaFileNames.begin()+i);
i--;
}
//make sure there is at least one valid file left
- if (fastaFileNames.size() == 0) { mothurOut("no valid files."); mothurOutEndLine(); abort = true; }
+ if (fastaFileNames.size() == 0) { m->mothurOut("no valid files."); m->mothurOutEndLine(); abort = true; }
}
taxonomyFileName = validParameter.validFile(parameters, "taxonomy", true);
if (taxonomyFileName == "not found") {
- mothurOut("taxonomy is a required parameter for the classify.seqs command.");
- mothurOutEndLine();
+ m->mothurOut("taxonomy is a required parameter for the classify.seqs command.");
+ m->mothurOutEndLine();
abort = true;
}
else if (taxonomyFileName == "not open") { abort = true; }
//go through files and make sure they are good, if not, then disregard them
for (int i = 0; i < namefileNames.size(); i++) {
+ if (inputDir != "") {
+ string path = hasPath(namefileNames[i]);
+ //if the user has not given a path then, add inputdir. else leave path alone.
+ if (path == "") { namefileNames[i] = inputDir + namefileNames[i]; }
+ }
+
int ableToOpen;
ifstream in;
ableToOpen = openInputFile(namefileNames[i], in);
- if (ableToOpen == 1) { mothurOut("Unable to match name file with fasta file."); mothurOutEndLine(); abort = true; }
+ if (ableToOpen == 1) { m->mothurOut("Unable to match name file with fasta file."); m->mothurOutEndLine(); abort = true; }
in.close();
}
}
if (namefile != "") {
- if (namefileNames.size() != fastaFileNames.size()) { abort = true; mothurOut("If you provide a name file, you must have one for each fasta file."); mothurOutEndLine(); }
+ 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 optional parameter and set defaults
if ((method == "bayesian") && (search != "kmer")) {
- mothurOut("The bayesian method requires the kmer search." + search + "will be disregarded." ); mothurOutEndLine();
+ m->mothurOut("The bayesian method requires the kmer search." + search + "will be disregarded." ); m->mothurOutEndLine();
search = "kmer";
}
}
}
catch(exception& e) {
- errorOut(e, "ClassifySeqsCommand", "ClassifySeqsCommand");
+ m->errorOut(e, "ClassifySeqsCommand", "ClassifySeqsCommand");
exit(1);
}
}
void ClassifySeqsCommand::help(){
try {
- mothurOut("The classify.seqs command reads a fasta file containing sequences and creates a .taxonomy file and a .tax.summary file.\n");
- mothurOut("The classify.seqs command parameters are template, fasta, search, ksize, method, taxonomy, processors, match, mismatch, gapopen, gapextend, numwanted and probs.\n");
- mothurOut("The template, 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");
- mothurOut("The search parameter allows you to specify the method to find most similar template. Your options are: suffix, kmer and blast. The default is kmer.\n");
- mothurOut("The method parameter allows you to specify classification method to use. Your options are: bayesian and knn. The default is bayesian.\n");
- mothurOut("The ksize parameter allows you to specify the kmer size for finding most similar template to candidate. The default is 8.\n");
- mothurOut("The processors parameter allows you to specify the number of processors to use. The default is 1.\n");
- mothurOut("The match parameter allows you to specify the bonus for having the same base. The default is 1.0.\n");
- mothurOut("The mistmatch parameter allows you to specify the penalty for having different bases. The default is -1.0.\n");
- mothurOut("The gapopen parameter allows you to specify the penalty for opening a gap in an alignment. The default is -2.0.\n");
- mothurOut("The gapextend parameter allows you to specify the penalty for extending a gap in an alignment. The default is -1.0.\n");
- mothurOut("The numwanted parameter allows you to specify the number of sequence matches you want with the knn method. The default is 10.\n");
- mothurOut("The cutoff parameter allows you to specify a bootstrap confidence threshold for your taxonomy. The default is 0.\n");
- mothurOut("The probs parameter shut off the bootstrapping results for the bayesian method. The default is true, meaning you want the bootstrapping to be run.\n");
- mothurOut("The iters parameter allows you to specify how many iterations to do when calculating the bootstrap confidence score for your taxonomy with the bayesian method. The default is 100.\n");
- mothurOut("The classify.seqs command should be in the following format: \n");
- mothurOut("classify.seqs(template=yourTemplateFile, fasta=yourFastaFile, method=yourClassificationMethod, search=yourSearchmethod, ksize=yourKmerSize, taxonomy=yourTaxonomyFile, processors=yourProcessors) \n");
- mothurOut("Example classify.seqs(fasta=amazon.fasta, template=core.filtered, method=knn, search=gotoh, ksize=8, processors=2)\n");
- mothurOut("The .taxonomy file consists of 2 columns: 1 = your sequence name, 2 = the taxonomy for your sequence. \n");
- mothurOut("The .tax.summary is a summary of the different taxonomies represented in your fasta file. \n");
- mothurOut("Note: No spaces between parameter labels (i.e. fasta), '=' and parameters (i.e.yourFastaFile).\n\n");
+ m->mothurOut("The classify.seqs command reads a fasta file containing sequences and creates a .taxonomy file and a .tax.summary file.\n");
+ m->mothurOut("The classify.seqs command parameters are template, fasta, name, search, ksize, method, taxonomy, processors, match, mismatch, gapopen, gapextend, numwanted and probs.\n");
+ m->mothurOut("The template, 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");
+ m->mothurOut("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");
+ m->mothurOut("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");
+ m->mothurOut("The method parameter allows you to specify classification method to use. Your options are: bayesian and knn. The default is bayesian.\n");
+ m->mothurOut("The ksize parameter allows you to specify the kmer size for finding most similar template to candidate. The default is 8.\n");
+ m->mothurOut("The processors parameter allows you to specify the number of processors to use. The default is 1.\n");
+ m->mothurOut("The match parameter allows you to specify the bonus for having the same base. The default is 1.0.\n");
+ m->mothurOut("The mistmatch parameter allows you to specify the penalty for having different bases. The default is -1.0.\n");
+ m->mothurOut("The gapopen parameter allows you to specify the penalty for opening a gap in an alignment. The default is -2.0.\n");
+ m->mothurOut("The gapextend parameter allows you to specify the penalty for extending a gap in an alignment. The default is -1.0.\n");
+ m->mothurOut("The numwanted parameter allows you to specify the number of sequence matches you want with the knn method. The default is 10.\n");
+ m->mothurOut("The cutoff parameter allows you to specify a bootstrap confidence threshold for your taxonomy. The default is 0.\n");
+ m->mothurOut("The probs parameter shut off the bootstrapping results for the bayesian method. The default is true, meaning you want the bootstrapping to be run.\n");
+ m->mothurOut("The iters parameter allows you to specify how many iterations to do when calculating the bootstrap confidence score for your taxonomy with the bayesian method. The default is 100.\n");
+ m->mothurOut("The classify.seqs command should be in the following format: \n");
+ m->mothurOut("classify.seqs(template=yourTemplateFile, fasta=yourFastaFile, method=yourClassificationMethod, search=yourSearchmethod, ksize=yourKmerSize, taxonomy=yourTaxonomyFile, processors=yourProcessors) \n");
+ m->mothurOut("Example classify.seqs(fasta=amazon.fasta, template=core.filtered, method=knn, search=gotoh, ksize=8, processors=2)\n");
+ m->mothurOut("The .taxonomy file consists of 2 columns: 1 = your sequence name, 2 = the taxonomy for your sequence. \n");
+ m->mothurOut("The .tax.summary is a summary of the different taxonomies represented in your fasta file. \n");
+ m->mothurOut("Note: No spaces between parameter labels (i.e. fasta), '=' and parameters (i.e.yourFastaFile).\n\n");
}
catch(exception& e) {
- errorOut(e, "ClassifySeqsCommand", "help");
+ m->errorOut(e, "ClassifySeqsCommand", "help");
exit(1);
}
}
if(method == "bayesian"){ classify = new Bayesian(taxonomyFileName, templateFileName, search, kmerSize, cutoff, iters); }
else if(method == "knn"){ classify = new Knn(taxonomyFileName, templateFileName, search, kmerSize, gapOpen, gapExtend, match, misMatch, numWanted); }
else {
- mothurOut(search + " is not a valid method option. I will run the command using bayesian.");
- mothurOutEndLine();
+ 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);
}
-
+
+ if (m->control_pressed) { delete classify; return 0; }
+
+ vector<string> outputNames;
for (int s = 0; s < fastaFileNames.size(); s++) {
inNames.close();
}
- mothurOut("Classifying sequences from " + fastaFileNames[s] + " ..." ); mothurOutEndLine();
- string newTaxonomyFile = getRootName(fastaFileNames[s]) + getRootName(taxonomyFileName) + "taxonomy";
- string tempTaxonomyFile = getRootName(fastaFileNames[s]) + "taxonomy.temp";
- string taxSummary = getRootName(fastaFileNames[s]) + getRootName(taxonomyFileName) + "tax.summary";
+ m->mothurOut("Classifying sequences from " + fastaFileNames[s] + " ..." ); m->mothurOutEndLine();
+
+ if (outputDir == "") { outputDir += hasPath(fastaFileNames[s]); }
+ string newTaxonomyFile = outputDir + getRootName(getSimpleName(fastaFileNames[s])) + getRootName(getSimpleName(taxonomyFileName)) + "taxonomy";
+ string tempTaxonomyFile = outputDir + getRootName(getSimpleName(fastaFileNames[s])) + "taxonomy.temp";
+ string taxSummary = outputDir + getRootName(getSimpleName(fastaFileNames[s])) + getRootName(getSimpleName(taxonomyFileName)) + "tax.summary";
+
+ outputNames.push_back(newTaxonomyFile);
+ outputNames.push_back(taxSummary);
int start = time(NULL);
int numFastaSeqs = 0;
#endif
//make taxonomy tree from new taxonomy file
PhyloTree taxaBrowser;
-
+
+ if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } delete classify; return 0; }
+
ifstream in;
openInputFile(tempTaxonomyFile, in);
while(!in.eof()){
in >> name >> taxon; gobble(in);
+ if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } remove(tempTaxonomyFile.c_str()); delete classify; return 0; }
+
if (namefile != "") {
itNames = nameMap.find(name);
if (itNames == nameMap.end()) {
- mothurOut(name + " is not in your name file please correct."); mothurOutEndLine(); exit(1);
+ m->mothurOut(name + " is not in your name file please correct."); m->mothurOutEndLine(); exit(1);
}else{
for (int i = 0; i < itNames->second; i++) {
taxaBrowser.addSeqToTree(name+toString(i), taxon); //add it as many times as there are identical seqs
in.close();
taxaBrowser.assignHeirarchyIDs(0);
+
+ if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } remove(tempTaxonomyFile.c_str()); delete classify; return 0; }
taxaBrowser.binUnclassified();
remove(tempTaxonomyFile.c_str());
+ if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } delete classify; return 0; }
+
+
//print summary file
ofstream outTaxTree;
openOutputFile(taxSummary, outTaxTree);
//get maxLevel from phylotree so you know how many 'unclassified's to add
int maxLevel = taxaBrowser.getMaxLevel();
- //read taxfile - this reading and rewriting is done to preserve the confidence sscores.
+ //read taxfile - this reading and rewriting is done to preserve the confidence scores.
while (!inTax.eof()) {
+ if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } remove(unclass.c_str()); delete classify; return 0; }
+
inTax >> name >> taxon; gobble(inTax);
string newTax = addUnclassifieds(taxon, maxLevel);
remove(newTaxonomyFile.c_str());
rename(unclass.c_str(), newTaxonomyFile.c_str());
- mothurOutEndLine();
- mothurOut("It took " + toString(time(NULL) - start) + " secs to classify " + toString(numFastaSeqs) + " sequences."); mothurOutEndLine(); mothurOutEndLine();
+ m->mothurOutEndLine();
+ m->mothurOut("Output File Names: "); m->mothurOutEndLine();
+ for (int i = 0; i < outputNames.size(); i++) { m->mothurOut(outputNames[i]); m->mothurOutEndLine(); }
+ m->mothurOutEndLine();
+
+
+ m->mothurOutEndLine();
+ m->mothurOut("It took " + toString(time(NULL) - start) + " secs to classify " + toString(numFastaSeqs) + " sequences."); m->mothurOutEndLine(); m->mothurOutEndLine();
}
delete classify;
return 0;
}
catch(exception& e) {
- errorOut(e, "ClassifySeqsCommand", "execute");
+ m->errorOut(e, "ClassifySeqsCommand", "execute");
exit(1);
}
}
return newTax;
}
catch(exception& e) {
- errorOut(e, "ClassifySeqsCommand", "addUnclassifieds");
+ m->errorOut(e, "ClassifySeqsCommand", "addUnclassifieds");
exit(1);
}
}
}else if (pid == 0){
driver(lines[process], taxFileName + toString(getpid()) + ".temp", tempTaxFile + toString(getpid()) + ".temp", filename);
exit(0);
- }else { mothurOut("unable to spawn the necessary processes."); mothurOutEndLine(); exit(0); }
+ }else { m->mothurOut("unable to spawn the necessary processes."); m->mothurOutEndLine(); exit(0); }
}
//force parent to wait until all the processes are done
#endif
}
catch(exception& e) {
- errorOut(e, "ClassifySeqsCommand", "createProcesses");
+ m->errorOut(e, "ClassifySeqsCommand", "createProcesses");
exit(1);
}
}
output.close();
}
catch(exception& e) {
- errorOut(e, "ClassifySeqsCommand", "appendTaxFiles");
+ m->errorOut(e, "ClassifySeqsCommand", "appendTaxFiles");
exit(1);
}
}
string taxonomy;
for(int i=0;i<line->numSeqs;i++){
+ if (m->control_pressed) { return 0; }
Sequence* candidateSeq = new Sequence(inFASTA);
if (candidateSeq->getName() != "") {
taxonomy = classify->getTaxonomy(candidateSeq);
+
+ if (m->control_pressed) { delete candidateSeq; return 0; }
if (taxonomy != "bad seq") {
//output confidence scores or not
outTax << candidateSeq->getName() << '\t' << taxonomy << endl;
}else{
outTax << candidateSeq->getName() << '\t' << classify->getSimpleTax() << endl;
- cout << classify->getSimpleTax() << endl;
}
outTaxSimple << candidateSeq->getName() << '\t' << classify->getSimpleTax() << endl;
delete candidateSeq;
if((i+1) % 100 == 0){
- mothurOut("Classifying sequence " + toString(i+1)); mothurOutEndLine();
+ m->mothurOut("Classifying sequence " + toString(i+1)); m->mothurOutEndLine();
}
}
return 1;
}
catch(exception& e) {
- errorOut(e, "ClassifySeqsCommand", "driver");
+ m->errorOut(e, "ClassifySeqsCommand", "driver");
exit(1);
}
}
projects / mothur.git / blobdiff