#include "phylosummary.h"
#include "knn.h"
-//**********************************************************************************************************************
+
+//**********************************************************************************************************************
+vector<string> ClassifySeqsCommand::setParameters(){
+ try {
+ 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 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 pprocessors("processors", "Number", "", "1", "", "", "",false,false); parameters.push_back(pprocessors);
+ CommandParameter pmatch("match", "Number", "", "1.0", "", "", "",false,false); parameters.push_back(pmatch);
+ CommandParameter pmismatch("mismatch", "Number", "", "-1.0", "", "", "",false,false); parameters.push_back(pmismatch);
+ CommandParameter pgapopen("gapopen", "Number", "", "-2.0", "", "", "",false,false); parameters.push_back(pgapopen);
+ CommandParameter pgapextend("gapextend", "Number", "", "-1.0", "", "", "",false,false); parameters.push_back(pgapextend);
+ CommandParameter pcutoff("cutoff", "Number", "", "0", "", "", "",false,true); parameters.push_back(pcutoff);
+ 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 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);
+
+ vector<string> myArray;
+ for (int i = 0; i < parameters.size(); i++) { myArray.push_back(parameters[i].name); }
+ return myArray;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "ClassifySeqsCommand", "setParameters");
+ exit(1);
+ }
+}
+//**********************************************************************************************************************
+string ClassifySeqsCommand::getHelpString(){
+ 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 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 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";
+#ifdef USE_MPI
+ helpString += "When using MPI, the processors parameter is set to the number of MPI processes running. \n";
+#endif
+ helpString += "If the save parameter is set to true the reference sequences will be saved in memory, to clear them later you can use the clear.memory command. Default=f.";
+ helpString += "The match parameter allows you to specify the bonus for having the same base. The default is 1.0.\n";
+ helpString += "The mistmatch parameter allows you to specify the penalty for having different bases. The default is -1.0.\n";
+ helpString += "The gapopen parameter allows you to specify the penalty for opening a gap in an alignment. The default is -2.0.\n";
+ helpString += "The gapextend parameter allows you to specify the penalty for extending a gap in an alignment. The default is -1.0.\n";
+ helpString += "The numwanted parameter allows you to specify the number of sequence matches you want with the knn method. The default is 10.\n";
+ helpString += "The cutoff parameter allows you to specify a bootstrap confidence threshold for your taxonomy. The default is 0.\n";
+ helpString += "The probs parameter shuts off the bootstrapping results for the bayesian method. The default is true, meaning you want the bootstrapping to be shown.\n";
+ helpString += "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";
+ helpString += "The classify.seqs command should be in the following format: \n";
+ helpString += "classify.seqs(reference=yourTemplateFile, fasta=yourFastaFile, method=yourClassificationMethod, search=yourSearchmethod, ksize=yourKmerSize, taxonomy=yourTaxonomyFile, processors=yourProcessors) \n";
+ helpString += "Example classify.seqs(fasta=amazon.fasta, reference=core.filtered, method=knn, search=gotoh, ksize=8, processors=2)\n";
+ helpString += "The .taxonomy file consists of 2 columns: 1 = your sequence name, 2 = the taxonomy for your sequence. \n";
+ helpString += "The .tax.summary is a summary of the different taxonomies represented in your fasta file. \n";
+ helpString += "Note: No spaces between parameter labels (i.e. fasta), '=' and parameters (i.e.yourFastaFile).\n";
+ return helpString;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "ClassifySeqsCommand", "getHelpString");
+ exit(1);
+ }
+}
+//**********************************************************************************************************************
+ClassifySeqsCommand::ClassifySeqsCommand(){
+ try {
+ abort = true; calledHelp = true;
+ setParameters();
+ vector<string> tempOutNames;
+ outputTypes["taxonomy"] = tempOutNames;
+ outputTypes["taxsummary"] = tempOutNames;
+ outputTypes["matchdist"] = tempOutNames;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "ClassifySeqsCommand", "ClassifySeqsCommand");
+ exit(1);
+ }
+}
+//**********************************************************************************************************************
ClassifySeqsCommand::ClassifySeqsCommand(string option) {
try {
- abort = false;
+ abort = false; calledHelp = false;
+ rdb = ReferenceDB::getInstance();
//allow user to run help
- if(option == "help") { help(); abort = true; }
+ if(option == "help") { help(); abort = true; calledHelp = true; }
+ else if(option == "citation") { citation(); abort = true; calledHelp = true;}
else {
-
- //valid paramters for this command
- string AlignArray[] = {"template","fasta","name","group","search","ksize","method","processors","taxonomy","match","mismatch","gapopen","gapextend","numwanted","cutoff","probs","iters", "outputdir","inputdir"};
- vector<string> myArray (AlignArray, AlignArray+(sizeof(AlignArray)/sizeof(string)));
+ vector<string> myArray = setParameters();
OptionParser parser(option);
map<string, string> parameters = parser.getParameters();
- ValidParameters validParameter;
+ ValidParameters validParameter("classify.seqs");
map<string, string>::iterator it;
//check to make sure all parameters are valid for command
if (validParameter.isValidParameter(it->first, myArray, it->second) != true) { abort = true; }
}
+ //initialize outputTypes
+ vector<string> tempOutNames;
+ outputTypes["taxonomy"] = tempOutNames;
+ outputTypes["taxsummary"] = tempOutNames;
+ outputTypes["matchdist"] = tempOutNames;
+
//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 (inputDir == "not found"){ inputDir = ""; }
else {
string path;
- it = parameters.find("template");
+ it = parameters.find("reference");
//user has given a template file
if(it != parameters.end()){
- path = hasPath(it->second);
+ path = m->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; }
+ if (path == "") { parameters["reference"] = inputDir + it->second; }
}
it = parameters.find("taxonomy");
//user has given a template file
if(it != parameters.end()){
- path = hasPath(it->second);
+ path = m->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; }
}
it = parameters.find("group");
//user has given a template file
if(it != parameters.end()){
- path = hasPath(it->second);
+ path = m->hasPath(it->second);
//if the user has not given a path then, add inputdir. else leave path alone.
if (path == "") { parameters["group"] = inputDir + it->second; }
}
}
- //check for required parameters
- templateFileName = validParameter.validFile(parameters, "template", true);
- if (templateFileName == "not found") {
- 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") { m->mothurOut("fasta is a required parameter for the classify.seqs command."); m->mothurOutEndLine(); abort = true; }
+ if (fastaFileName == "not found") {
+ //if there is a current fasta file, use it
+ string filename = m->getFastaFile();
+ if (filename != "") { fastaFileNames.push_back(filename); m->mothurOut("Using " + filename + " as input file for the fasta parameter."); m->mothurOutEndLine(); }
+ else { m->mothurOut("You have no current fastafile and the fasta parameter is required."); m->mothurOutEndLine(); abort = true; }
+ }
else {
- splitAtDash(fastaFileName, fastaFileNames);
+ m->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;
- #ifdef USE_MPI
- int pid;
- MPI_Comm_size(MPI_COMM_WORLD, &processors); //set processors to the number of mpi processes running
- MPI_Comm_rank(MPI_COMM_WORLD, &pid); //find out who we are
-
- if (pid == 0) {
- #endif
+ bool ignore = false;
+ if (fastaFileNames[i] == "current") {
+ fastaFileNames[i] = m->getFastaFile();
+ if (fastaFileNames[i] != "") { m->mothurOut("Using " + fastaFileNames[i] + " as input file for the fasta parameter where you had given current."); m->mothurOutEndLine(); }
+ else {
+ m->mothurOut("You have no current fastafile, ignoring current."); m->mothurOutEndLine(); ignore=true;
+ //erase from file list
+ fastaFileNames.erase(fastaFileNames.begin()+i);
+ i--;
+ }
+ }
- ifstream in;
- ableToOpen = openInputFile(fastaFileNames[i], in);
- in.close();
+ if (!ignore) {
+
+ if (inputDir != "") {
+ string path = m->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 = m->openInputFile(fastaFileNames[i], in, "noerror");
- #ifdef USE_MPI
- for (int j = 1; j < processors; j++) {
- MPI_Send(&ableToOpen, 1, MPI_INT, j, 2001, MPI_COMM_WORLD);
+ //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(fastaFileNames[i]);
+ m->mothurOut("Unable to open " + fastaFileNames[i] + ". Trying default " + tryPath); m->mothurOutEndLine();
+ ifstream in2;
+ ableToOpen = m->openInputFile(tryPath, in2, "noerror");
+ in2.close();
+ fastaFileNames[i] = tryPath;
}
- }else{
- MPI_Status status;
- MPI_Recv(&ableToOpen, 1, MPI_INT, 0, 2001, MPI_COMM_WORLD, &status);
}
- #endif
-
- if (ableToOpen == 1) {
- m->mothurOut(fastaFileNames[i] + " will be disregarded."); m->mothurOutEndLine();
- //erase from file list
- fastaFileNames.erase(fastaFileNames.begin()+i);
- i--;
+ if (ableToOpen == 1) {
+ if (m->getOutputDir() != "") { //default path is set
+ string tryPath = m->getOutputDir() + m->getSimpleName(fastaFileNames[i]);
+ m->mothurOut("Unable to open " + fastaFileNames[i] + ". Trying output directory " + tryPath); m->mothurOutEndLine();
+ ifstream in2;
+ ableToOpen = m->openInputFile(tryPath, in2, "noerror");
+ in2.close();
+ fastaFileNames[i] = tryPath;
+ }
+ }
+
+ in.close();
+
+ if (ableToOpen == 1) {
+ m->mothurOut("Unable to open " + fastaFileNames[i] + ". It will be disregarded."); m->mothurOutEndLine();
+ //erase from file list
+ fastaFileNames.erase(fastaFileNames.begin()+i);
+ i--;
+ }else {
+ m->setFastaFile(fastaFileNames[i]);
+ }
}
}
if (fastaFileNames.size() == 0) { m->mothurOut("no valid files."); m->mothurOutEndLine(); abort = true; }
}
-
- taxonomyFileName = validParameter.validFile(parameters, "taxonomy", true);
- if (taxonomyFileName == "not found") {
- m->mothurOut("taxonomy is a required parameter for the classify.seqs command.");
- m->mothurOutEndLine();
- abort = true;
- }
- else if (taxonomyFileName == "not open") { abort = true; }
-
-
namefile = validParameter.validFile(parameters, "name", false);
if (namefile == "not found") { namefile = ""; }
else {
- splitAtDash(namefile, namefileNames);
+ m->splitAtDash(namefile, namefileNames);
//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]; }
+ bool ignore = false;
+ if (namefileNames[i] == "current") {
+ namefileNames[i] = m->getNameFile();
+ if (namefileNames[i] != "") { m->mothurOut("Using " + namefileNames[i] + " as input file for the name parameter where you had given current."); m->mothurOutEndLine(); }
+ else {
+ m->mothurOut("You have no current namefile, ignoring current."); m->mothurOutEndLine(); ignore=true;
+ //erase from file list
+ namefileNames.erase(namefileNames.begin()+i);
+ i--;
+ }
}
- int ableToOpen;
- #ifdef USE_MPI
- int pid;
- MPI_Comm_size(MPI_COMM_WORLD, &processors); //set processors to the number of mpi processes running
- MPI_Comm_rank(MPI_COMM_WORLD, &pid); //find out who we are
-
- if (pid == 0) {
- #endif
-
- ifstream in;
- ableToOpen = openInputFile(namefileNames[i], in);
- in.close();
+ if (!ignore) {
+
+ if (inputDir != "") {
+ string path = m->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 = m->openInputFile(namefileNames[i], in, "noerror");
- #ifdef USE_MPI
- for (int j = 1; j < processors; j++) {
- MPI_Send(&ableToOpen, 1, MPI_INT, j, 2001, MPI_COMM_WORLD);
+ //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(namefileNames[i]);
+ m->mothurOut("Unable to open " + namefileNames[i] + ". Trying default " + tryPath); m->mothurOutEndLine();
+ ifstream in2;
+ ableToOpen = m->openInputFile(tryPath, in2, "noerror");
+ in2.close();
+ namefileNames[i] = tryPath;
}
- }else{
- MPI_Status status;
- MPI_Recv(&ableToOpen, 1, MPI_INT, 0, 2001, MPI_COMM_WORLD, &status);
}
- #endif
- if (ableToOpen == 1) { m->mothurOut("Unable to match name file with fasta file."); m->mothurOutEndLine(); abort = true; }
-
+ if (ableToOpen == 1) {
+ if (m->getOutputDir() != "") { //default path is set
+ string tryPath = m->getOutputDir() + m->getSimpleName(namefileNames[i]);
+ m->mothurOut("Unable to open " + namefileNames[i] + ". Trying output directory " + tryPath); m->mothurOutEndLine();
+ ifstream in2;
+ ableToOpen = m->openInputFile(tryPath, in2, "noerror");
+ in2.close();
+ namefileNames[i] = tryPath;
+ }
+ }
+ in.close();
+
+ if (ableToOpen == 1) {
+ m->mothurOut("Unable to open " + namefileNames[i] + ". It will be disregarded."); m->mothurOutEndLine(); abort = true;
+ //erase from file list
+ namefileNames.erase(namefileNames.begin()+i);
+ i--;
+ }else {
+ m->setNameFile(namefileNames[i]);
+ }
+ }
}
}
groupfile = validParameter.validFile(parameters, "group", false);
if (groupfile == "not found") { groupfile = ""; }
else {
- splitAtDash(groupfile, groupfileNames);
+ m->splitAtDash(groupfile, groupfileNames);
//go through files and make sure they are good, if not, then disregard them
for (int i = 0; i < groupfileNames.size(); i++) {
if (inputDir != "") {
- string path = hasPath(groupfileNames[i]);
+ string path = m->hasPath(groupfileNames[i]);
//if the user has not given a path then, add inputdir. else leave path alone.
if (path == "") { groupfileNames[i] = inputDir + groupfileNames[i]; }
}
int ableToOpen;
- #ifdef USE_MPI
- int pid;
- MPI_Comm_size(MPI_COMM_WORLD, &processors); //set processors to the number of mpi processes running
- MPI_Comm_rank(MPI_COMM_WORLD, &pid); //find out who we are
-
- if (pid == 0) {
- #endif
-
ifstream in;
- ableToOpen = openInputFile(groupfileNames[i], in);
- in.close();
+ ableToOpen = m->openInputFile(groupfileNames[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(groupfileNames[i]);
+ m->mothurOut("Unable to open " + groupfileNames[i] + ". Trying default " + tryPath); m->mothurOutEndLine();
+ ifstream in2;
+ ableToOpen = m->openInputFile(tryPath, in2, "noerror");
+ in2.close();
+ groupfileNames[i] = tryPath;
+ }
+ }
- #ifdef USE_MPI
- for (int j = 1; j < processors; j++) {
- MPI_Send(&ableToOpen, 1, MPI_INT, j, 2001, MPI_COMM_WORLD);
- }
- }else{
- MPI_Status status;
- MPI_Recv(&ableToOpen, 1, MPI_INT, 0, 2001, MPI_COMM_WORLD, &status);
+ if (ableToOpen == 1) {
+ if (m->getOutputDir() != "") { //default path is set
+ string tryPath = m->getOutputDir() + m->getSimpleName(groupfileNames[i]);
+ m->mothurOut("Unable to open " + groupfileNames[i] + ". Trying output directory " + tryPath); m->mothurOutEndLine();
+ ifstream in2;
+ ableToOpen = m->openInputFile(tryPath, in2, "noerror");
+ in2.close();
+ groupfileNames[i] = tryPath;
}
-
- #endif
- if (ableToOpen == 1) { m->mothurOut("Unable to match group file with fasta file."); m->mothurOutEndLine(); abort = true; }
+ }
+ in.close();
+
+ if (ableToOpen == 1) {
+ m->mothurOut("Unable to open " + groupfileNames[i] + ". It will be disregarded."); m->mothurOutEndLine(); groupfileNames[i] = "";
+ //erase from file list
+ groupfileNames.erase(groupfileNames.begin()+i);
+ i--;
+ }else {
+ m->setGroupFile(groupfileNames[i]);
+ }
}
}
temp = validParameter.validFile(parameters, "ksize", false); if (temp == "not found"){ temp = "8"; }
convert(temp, kmerSize);
- temp = validParameter.validFile(parameters, "processors", false); if (temp == "not found"){ temp = "1"; }
+ temp = validParameter.validFile(parameters, "save", false); if (temp == "not found"){ temp = "f"; }
+ save = m->isTrue(temp);
+ rdb->save = save;
+ if (save) { //clear out old references
+ rdb->clearMemory();
+ }
+
+ //this has to go after save so that if the user sets save=t and provides no reference we abort
+ templateFileName = validParameter.validFile(parameters, "reference", true);
+ if (templateFileName == "not found") {
+ //check for saved reference sequences
+ if (rdb->referenceSeqs.size() != 0) {
+ templateFileName = "saved";
+ }else {
+ m->mothurOut("[ERROR]: You don't have any saved reference sequences and the reference parameter is a required for the classify.seqs command.");
+ m->mothurOutEndLine();
+ abort = true;
+ }
+ }else if (templateFileName == "not open") { abort = true; }
+ else { if (save) { rdb->setSavedReference(templateFileName); } }
+
+ //this has to go after save so that if the user sets save=t and provides no reference we abort
+ taxonomyFileName = validParameter.validFile(parameters, "taxonomy", true);
+ if (taxonomyFileName == "not found") {
+ //check for saved reference sequences
+ if (rdb->wordGenusProb.size() != 0) {
+ taxonomyFileName = "saved";
+ }else {
+ m->mothurOut("[ERROR]: You don't have any saved taxonomy information and the taxonomy parameter is a required for the classify.seqs command.");
+ m->mothurOutEndLine();
+ abort = true;
+ }
+ }else if (taxonomyFileName == "not open") { abort = true; }
+ else { if (save) { rdb->setSavedTaxonomy(taxonomyFileName); } }
+
+ temp = validParameter.validFile(parameters, "processors", false); if (temp == "not found"){ temp = m->getProcessors(); }
+ m->setProcessors(temp);
convert(temp, processors);
search = validParameter.validFile(parameters, "search", false); if (search == "not found"){ search = "kmer"; }
convert(temp, cutoff);
temp = validParameter.validFile(parameters, "probs", false); if (temp == "not found"){ temp = "true"; }
- probs = isTrue(temp);
+ probs = m->isTrue(temp);
temp = validParameter.validFile(parameters, "iters", false); if (temp == "not found") { temp = "100"; }
convert(temp, iters);
}
//**********************************************************************************************************************
-
ClassifySeqsCommand::~ClassifySeqsCommand(){
-
if (abort == false) {
for (int i = 0; i < lines.size(); i++) { delete lines[i]; } lines.clear();
}
}
-
-//**********************************************************************************************************************
-
-void ClassifySeqsCommand::help(){
- try {
- 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 group parameter allows you add a group file so you can have the summary totals broken up by group.\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");
- #ifdef USE_MPI
- m->mothurOut("When using MPI, the processors parameter is set to the number of MPI processes running. \n");
- #endif
- 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) {
- m->errorOut(e, "ClassifySeqsCommand", "help");
- exit(1);
- }
-}
-
-
//**********************************************************************************************************************
int ClassifySeqsCommand::execute(){
try {
- if (abort == true) { return 0; }
+ if (abort == true) { if (calledHelp) { return 0; } return 2; }
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); }
if (m->control_pressed) { delete classify; return 0; }
- vector<string> outputNames;
for (int s = 0; s < fastaFileNames.size(); s++) {
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";
+ string baseTName = taxonomyFileName;
+ if (taxonomyFileName == "saved") {baseTName = rdb->getSavedTaxonomy(); }
- outputNames.push_back(newTaxonomyFile);
- outputNames.push_back(taxSummary);
+ 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 += ".";
+
+ if (outputDir == "") { outputDir += m->hasPath(fastaFileNames[s]); }
+ string newTaxonomyFile = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + RippedTaxName + "taxonomy";
+ string tempTaxonomyFile = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + "taxonomy.temp";
+ string taxSummary = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + RippedTaxName + "tax.summary";
+
+ if ((method == "knn") && (search == "distance")) {
+ string DistName = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + "match.dist";
+ classify->setDistName(DistName); outputNames.push_back(DistName); outputTypes["matchdist"].push_back(DistName);
+ }
+
+ outputNames.push_back(newTaxonomyFile); outputTypes["taxonomy"].push_back(newTaxonomyFile);
+ outputNames.push_back(taxSummary); outputTypes["taxsummary"].push_back(taxSummary);
int start = time(NULL);
int numFastaSeqs = 0;
for (int i = 0; i < lines.size(); i++) { delete lines[i]; } lines.clear();
#ifdef USE_MPI
- int pid, end, numSeqsPerProcessor;
+ int pid, numSeqsPerProcessor;
int tag = 2001;
- vector<long> MPIPos;
+ vector<unsigned long int> MPIPos;
MPI_Status status;
MPI_Comm_rank(MPI_COMM_WORLD, &pid); //find out who we are
int outMode=MPI_MODE_CREATE|MPI_MODE_WRONLY;
int inMode=MPI_MODE_RDONLY;
- //char* outNewTax = new char[newTaxonomyFile.length()];
- //memcpy(outNewTax, newTaxonomyFile.c_str(), newTaxonomyFile.length());
-
char outNewTax[1024];
strcpy(outNewTax, newTaxonomyFile.c_str());
-
- //char* outTempTax = new char[tempTaxonomyFile.length()];
- //memcpy(outTempTax, tempTaxonomyFile.c_str(), tempTaxonomyFile.length());
char outTempTax[1024];
strcpy(outTempTax, tempTaxonomyFile.c_str());
-
- //char* inFileName = new char[fastaFileNames[s].length()];
- //memcpy(inFileName, fastaFileNames[s].c_str(), fastaFileNames[s].length());
char inFileName[1024];
strcpy(inFileName, fastaFileNames[s].c_str());
MPI_File_open(MPI_COMM_WORLD, outNewTax, outMode, MPI_INFO_NULL, &outMPINewTax);
MPI_File_open(MPI_COMM_WORLD, outTempTax, outMode, MPI_INFO_NULL, &outMPITempTax);
- //delete outNewTax;
- //delete outTempTax;
- //delete inFileName;
-
- if (m->control_pressed) { MPI_File_close(&inMPI); MPI_File_close(&outMPINewTax); MPI_File_close(&outMPITempTax); delete classify; return 0; }
-
- if(namefile != "") { MPIReadNamesFile(namefileNames[s]); }
+ if (m->control_pressed) { outputTypes.clear(); MPI_File_close(&inMPI); MPI_File_close(&outMPINewTax); MPI_File_close(&outMPITempTax); delete classify; return 0; }
if (pid == 0) { //you are the root process
- MPIPos = setFilePosFasta(fastaFileNames[s], numFastaSeqs); //fills MPIPos, returns numSeqs
+ MPIPos = m->setFilePosFasta(fastaFileNames[s], numFastaSeqs); //fills MPIPos, returns numSeqs
//send file positions to all processes
- MPI_Bcast(&numFastaSeqs, 1, MPI_INT, 0, MPI_COMM_WORLD); //send numSeqs
- MPI_Bcast(&MPIPos[0], (numFastaSeqs+1), MPI_LONG, 0, MPI_COMM_WORLD); //send file pos
+ for(int i = 1; i < processors; i++) {
+ MPI_Send(&numFastaSeqs, 1, MPI_INT, i, tag, MPI_COMM_WORLD);
+ MPI_Send(&MPIPos[0], (numFastaSeqs+1), MPI_LONG, i, tag, MPI_COMM_WORLD);
+ }
//figure out how many sequences you have to align
numSeqsPerProcessor = numFastaSeqs / processors;
//align your part
driverMPI(startIndex, numSeqsPerProcessor, inMPI, outMPINewTax, outMPITempTax, MPIPos);
- if (m->control_pressed) { MPI_File_close(&inMPI); MPI_File_close(&outMPINewTax); MPI_File_close(&outMPITempTax); for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } delete classify; return 0; }
+ if (m->control_pressed) { outputTypes.clear(); MPI_File_close(&inMPI); MPI_File_close(&outMPINewTax); MPI_File_close(&outMPITempTax); for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } delete classify; return 0; }
for (int i = 1; i < processors; i++) {
int done;
MPI_Recv(&done, 1, MPI_INT, i, tag, MPI_COMM_WORLD, &status);
}
}else{ //you are a child process
- MPI_Bcast(&numFastaSeqs, 1, MPI_INT, 0, MPI_COMM_WORLD); //get numSeqs
+ MPI_Recv(&numFastaSeqs, 1, MPI_INT, 0, tag, MPI_COMM_WORLD, &status);
MPIPos.resize(numFastaSeqs+1);
- MPI_Bcast(&MPIPos[0], (numFastaSeqs+1), MPI_LONG, 0, MPI_COMM_WORLD); //get file positions
+ MPI_Recv(&MPIPos[0], (numFastaSeqs+1), MPI_LONG, 0, tag, MPI_COMM_WORLD, &status);
//figure out how many sequences you have to align
numSeqsPerProcessor = numFastaSeqs / processors;
//align your part
driverMPI(startIndex, numSeqsPerProcessor, inMPI, outMPINewTax, outMPITempTax, MPIPos);
- if (m->control_pressed) { MPI_File_close(&inMPI); MPI_File_close(&outMPINewTax); MPI_File_close(&outMPITempTax); delete classify; return 0; }
+ if (m->control_pressed) { outputTypes.clear(); MPI_File_close(&inMPI); MPI_File_close(&outMPINewTax); MPI_File_close(&outMPITempTax); delete classify; return 0; }
int done = 0;
MPI_Send(&done, 1, MPI_INT, 0, tag, MPI_COMM_WORLD);
MPI_File_close(&inMPI);
MPI_File_close(&outMPINewTax);
MPI_File_close(&outMPITempTax);
+ MPI_Barrier(MPI_COMM_WORLD); //make everyone wait - just in case
#else
- //read namefile
- if(namefile != "") {
- nameMap.clear(); //remove old names
-
- ifstream inNames;
- openInputFile(namefileNames[s], inNames);
+
+ vector<unsigned long int> positions = m->divideFile(fastaFileNames[s], processors);
- string firstCol, secondCol;
- while(!inNames.eof()) {
- inNames >> firstCol >> secondCol; gobble(inNames);
- nameMap[firstCol] = getNumNames(secondCol); //ex. seq1 seq1,seq3,seq5 -> seq1 = 3.
- }
- inNames.close();
- }
-
- #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
+ for (int i = 0; i < (positions.size()-1); i++) {
+ lines.push_back(new linePair(positions[i], positions[(i+1)]));
+ }
+
+ #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
if(processors == 1){
- ifstream inFASTA;
- openInputFile(fastaFileNames[s], inFASTA);
- numFastaSeqs=count(istreambuf_iterator<char>(inFASTA),istreambuf_iterator<char>(), '>');
- inFASTA.close();
-
- lines.push_back(new linePair(0, numFastaSeqs));
-
- driver(lines[0], newTaxonomyFile, tempTaxonomyFile, fastaFileNames[s]);
+ numFastaSeqs = driver(lines[0], newTaxonomyFile, tempTaxonomyFile, fastaFileNames[s]);
}
else{
- vector<int> positions;
processIDS.resize(0);
- ifstream inFASTA;
- openInputFile(fastaFileNames[s], inFASTA);
-
- string input;
- while(!inFASTA.eof()){
- input = getline(inFASTA);
- if (input.length() != 0) {
- if(input[0] == '>'){ int pos = inFASTA.tellg(); positions.push_back(pos - input.length() - 1); }
- }
- }
- inFASTA.close();
-
- numFastaSeqs = positions.size();
-
- int numSeqsPerProcessor = numFastaSeqs / processors;
-
- for (int i = 0; i < processors; i++) {
- int startPos = positions[ i * numSeqsPerProcessor ];
- if(i == processors - 1){
- numSeqsPerProcessor = numFastaSeqs - i * numSeqsPerProcessor;
- }
- lines.push_back(new linePair(startPos, numSeqsPerProcessor));
- }
- createProcesses(newTaxonomyFile, tempTaxonomyFile, fastaFileNames[s]);
-
- rename((newTaxonomyFile + toString(processIDS[0]) + ".temp").c_str(), newTaxonomyFile.c_str());
- rename((tempTaxonomyFile + toString(processIDS[0]) + ".temp").c_str(), tempTaxonomyFile.c_str());
-
- for(int i=1;i<processors;i++){
- appendTaxFiles((newTaxonomyFile + toString(processIDS[i]) + ".temp"), newTaxonomyFile);
- appendTaxFiles((tempTaxonomyFile + toString(processIDS[i]) + ".temp"), tempTaxonomyFile);
- remove((newTaxonomyFile + toString(processIDS[i]) + ".temp").c_str());
- remove((tempTaxonomyFile + toString(processIDS[i]) + ".temp").c_str());
- }
+ numFastaSeqs = createProcesses(newTaxonomyFile, tempTaxonomyFile, fastaFileNames[s]);
}
#else
- ifstream inFASTA;
- openInputFile(fastaFileNames[s], inFASTA);
- numFastaSeqs=count(istreambuf_iterator<char>(inFASTA),istreambuf_iterator<char>(), '>');
- inFASTA.close();
-
- lines.push_back(new linePair(0, numFastaSeqs));
-
- driver(lines[0], newTaxonomyFile, tempTaxonomyFile, fastaFileNames[s]);
+ numFastaSeqs = driver(lines[0], newTaxonomyFile, tempTaxonomyFile, fastaFileNames[s]);
#endif
#endif
+ m->mothurOutEndLine();
+ m->mothurOut("It took " + toString(time(NULL) - start) + " secs to classify " + toString(numFastaSeqs) + " sequences."); m->mothurOutEndLine(); m->mothurOutEndLine();
+ start = time(NULL);
+
+
#ifdef USE_MPI
if (pid == 0) { //this part does not need to be paralellized
+
+ if(namefile != "") { m->mothurOut("Reading " + namefileNames[s] + "..."); cout.flush(); MPIReadNamesFile(namefileNames[s]); m->mothurOut(" Done."); m->mothurOutEndLine(); }
+ #else
+ //read namefile
+ 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->mothurOut(" Done."); m->mothurOutEndLine();
+ }
#endif
- m->mothurOutEndLine();
- m->mothurOut("It took " + toString(time(NULL) - start) + " secs to classify " + toString(numFastaSeqs) + " sequences."); m->mothurOutEndLine(); m->mothurOutEndLine();
- start = time(NULL);
-
- PhyloSummary taxaSum(taxonomyFileName, groupfileNames[s]);
+ string group = "";
+ if (groupfile != "") { group = groupfileNames[s]; }
- if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } delete classify; return 0; }
+ 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;
- openInputFile(tempTaxonomyFile, in);
+ m->openInputFile(tempTaxonomyFile, in);
//read in users taxonomy file and add sequences to tree
string name, taxon;
+
while(!in.eof()){
- in >> name >> taxon; gobble(in);
+ 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; i++) {
- taxaSum.addSeqToTree(name, taxon); //add it as many times as there are identical seqs
+ 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();
}
- remove(tempTaxonomyFile.c_str());
+ m->mothurRemove(tempTaxonomyFile);
- if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } delete classify; return 0; }
+ if (m->control_pressed) { outputTypes.clear(); for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } delete classify; return 0; }
//print summary file
ofstream outTaxTree;
- openOutputFile(taxSummary, outTaxTree);
+ m->openOutputFile(taxSummary, outTaxTree);
taxaSum.print(outTaxTree);
outTaxTree.close();
//output taxonomy with the unclassified bins added
ifstream inTax;
- openInputFile(newTaxonomyFile, inTax);
+ m->openInputFile(newTaxonomyFile, inTax);
ofstream outTax;
string unclass = newTaxonomyFile + ".unclass.temp";
- openOutputFile(unclass, outTax);
+ m->openOutputFile(unclass, outTax);
//get maxLevel from phylotree so you know how many 'unclassified's to add
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) { for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } remove(unclass.c_str()); delete classify; return 0; }
+ 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; }
- inTax >> name >> taxon; gobble(inTax);
+ inTax >> name >> taxon; m->gobble(inTax);
string newTax = addUnclassifieds(taxon, maxLevel);
inTax.close();
outTax.close();
- remove(newTaxonomyFile.c_str());
+ m->mothurRemove(newTaxonomyFile);
rename(unclass.c_str(), newTaxonomyFile.c_str());
m->mothurOutEndLine();
- m->mothurOut("It took " + toString(time(NULL) - start) + " secs to create the summary file for " + toString(numFastaSeqs) + " sequences."); m->mothurOutEndLine(); m->mothurOutEndLine();
+ m->mothurOut("It took " + toString(time(NULL) - start) + " secs to create the summary file for " + toString(numFastaSeqs) + " sequences."); m->mothurOutEndLine(); m->mothurOutEndLine();
#ifdef USE_MPI
}
m->mothurOutEndLine();
}
+ //set taxonomy file as new current taxonomyfile
+ string current = "";
+ itTypes = outputTypes.find("taxonomy");
+ if (itTypes != outputTypes.end()) {
+ if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setTaxonomyFile(current); }
+ }
+
delete classify;
return 0;
}
/**************************************************************************************************/
-void ClassifySeqsCommand::createProcesses(string taxFileName, string tempTaxFile, string filename) {
+int ClassifySeqsCommand::createProcesses(string taxFileName, string tempTaxFile, string filename) {
try {
#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
- int process = 0;
- // processIDS.resize(0);
+ int process = 1;
+ int num = 0;
//loop through and create all the processes you want
while (process != processors) {
processIDS.push_back(pid); //create map from line number to pid so you can append files in correct order later
process++;
}else if (pid == 0){
- driver(lines[process], taxFileName + toString(getpid()) + ".temp", tempTaxFile + toString(getpid()) + ".temp", filename);
+ num = driver(lines[process], taxFileName + toString(getpid()) + ".temp", tempTaxFile + toString(getpid()) + ".temp", filename);
+
+ //pass numSeqs to parent
+ ofstream out;
+ string tempFile = filename + toString(getpid()) + ".num.temp";
+ m->openOutputFile(tempFile, out);
+ out << num << endl;
+ out.close();
+
+ exit(0);
+ }else {
+ m->mothurOut("[ERROR]: unable to spawn the necessary processes."); m->mothurOutEndLine();
+ for (int i = 0; i < processIDS.size(); i++) { kill (processIDS[i], SIGINT); }
exit(0);
- }else { m->mothurOut("unable to spawn the necessary processes."); m->mothurOutEndLine(); exit(0); }
+ }
}
+ //parent does its part
+ num = driver(lines[0], taxFileName, tempTaxFile, filename);
+
//force parent to wait until all the processes are done
- for (int i=0;i<processors;i++) {
+ for (int i=0;i<processIDS.size();i++) {
int temp = processIDS[i];
wait(&temp);
}
+
+ for (int i = 0; i < processIDS.size(); i++) {
+ ifstream in;
+ string tempFile = filename + toString(processIDS[i]) + ".num.temp";
+ m->openInputFile(tempFile, in);
+ if (!in.eof()) { int tempNum = 0; in >> tempNum; num += tempNum; }
+ in.close(); m->mothurRemove(m->getFullPathName(tempFile));
+ }
+
+ for(int i=0;i<processIDS.size();i++){
+ appendTaxFiles((taxFileName + toString(processIDS[i]) + ".temp"), taxFileName);
+ appendTaxFiles((tempTaxFile + toString(processIDS[i]) + ".temp"), tempTaxFile);
+ m->mothurRemove((m->getFullPathName(taxFileName) + toString(processIDS[i]) + ".temp"));
+ m->mothurRemove((m->getFullPathName(tempTaxFile) + toString(processIDS[i]) + ".temp"));
+ }
+
+ return num;
#endif
}
catch(exception& e) {
ofstream output;
ifstream input;
- openOutputFileAppend(filename, output);
- openInputFile(temp, input);
+ m->openOutputFileAppend(filename, output);
+ m->openInputFile(temp, input);
while(char c = input.get()){
if(input.eof()) { break; }
//**********************************************************************************************************************
-int ClassifySeqsCommand::driver(linePair* line, string taxFName, string tempTFName, string filename){
+int ClassifySeqsCommand::driver(linePair* filePos, string taxFName, string tempTFName, string filename){
try {
ofstream outTax;
- openOutputFile(taxFName, outTax);
+ m->openOutputFile(taxFName, outTax);
ofstream outTaxSimple;
- openOutputFile(tempTFName, outTaxSimple);
+ m->openOutputFile(tempTFName, outTaxSimple);
ifstream inFASTA;
- openInputFile(filename, inFASTA);
-
- inFASTA.seekg(line->start);
+ m->openInputFile(filename, inFASTA);
string taxonomy;
- for(int i=0;i<line->numSeqs;i++){
+ inFASTA.seekg(filePos->start);
+
+ bool done = false;
+ int count = 0;
+
+ while (!done) {
if (m->control_pressed) { return 0; }
-
- Sequence* candidateSeq = new Sequence(inFASTA);
+
+ Sequence* candidateSeq = new Sequence(inFASTA); m->gobble(inFASTA);
if (candidateSeq->getName() != "") {
+
taxonomy = classify->getTaxonomy(candidateSeq);
if (m->control_pressed) { delete candidateSeq; return 0; }
outTaxSimple << candidateSeq->getName() << '\t' << classify->getSimpleTax() << endl;
}
- }
+ count++;
+ }
delete candidateSeq;
- if((i+1) % 100 == 0){
- m->mothurOut("Classifying sequence " + toString(i+1)); m->mothurOutEndLine();
- }
+ #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
+ unsigned long int pos = inFASTA.tellg();
+ if ((pos == -1) || (pos >= filePos->end)) { break; }
+ #else
+ if (inFASTA.eof()) { break; }
+ #endif
+
+ //report progress
+ if((count) % 100 == 0){ m->mothurOut("Processing sequence: " + toString(count)); m->mothurOutEndLine(); }
+
}
-
+ //report progress
+ if((count) % 100 != 0){ m->mothurOut("Processing sequence: " + toString(count)); m->mothurOutEndLine(); }
+
inFASTA.close();
outTax.close();
outTaxSimple.close();
- return 1;
+ return count;
}
catch(exception& e) {
m->errorOut(e, "ClassifySeqsCommand", "driver");
}
//**********************************************************************************************************************
#ifdef USE_MPI
-int ClassifySeqsCommand::driverMPI(int start, int num, MPI_File& inMPI, MPI_File& newFile, MPI_File& tempFile, vector<long>& MPIPos){
+int ClassifySeqsCommand::driverMPI(int start, int num, MPI_File& inMPI, MPI_File& newFile, MPI_File& tempFile, vector<unsigned long int>& MPIPos){
try {
MPI_Status statusNew;
MPI_Status statusTemp;
string firstCol, secondCol;
while(!iss.eof()) {
- iss >> firstCol >> secondCol; gobble(iss);
- nameMap[firstCol] = getNumNames(secondCol); //ex. seq1 seq1,seq3,seq5 -> seq1 = 3.
+ iss >> firstCol >> secondCol; m->gobble(iss);
+
+ vector<string> temp;
+ m->splitAtComma(secondCol, temp);
+
+ nameMap[firstCol] = temp;
}
MPI_File_close(&inMPI);
projects / mothur.git / blobdiff