*/
#include "treegroupscommand.h"
-#include "sharedjabund.h"
-#include "sharedsorabund.h"
-#include "sharedjclass.h"
-#include "sharedsorclass.h"
-#include "sharedjest.h"
-#include "sharedsorest.h"
-#include "sharedthetayc.h"
-#include "sharedthetan.h"
-#include "sharedmorisitahorn.h"
-#include "sharedbraycurtis.h"
-
+#include "subsample.h"
+#include "consensus.h"
+//**********************************************************************************************************************
+vector<string> TreeGroupCommand::setParameters(){
+ try {
+ CommandParameter pshared("shared", "InputTypes", "", "", "PhylipColumnShared", "PhylipColumnShared", "none",false,false); parameters.push_back(pshared);
+ CommandParameter pphylip("phylip", "InputTypes", "", "", "PhylipColumnShared", "PhylipColumnShared", "none",false,false); parameters.push_back(pphylip);
+ CommandParameter pname("name", "InputTypes", "", "", "NameCount", "none", "ColumnName",false,false); parameters.push_back(pname);
+ CommandParameter pcount("count", "InputTypes", "", "", "NameCount", "none", "countcolumn",false,false); parameters.push_back(pcount);
+ CommandParameter pcolumn("column", "InputTypes", "", "", "PhylipColumnShared", "PhylipColumnShared", "ColumnName-countcolumn",false,false); parameters.push_back(pcolumn);
+ CommandParameter piters("iters", "Number", "", "1000", "", "", "",false,false); parameters.push_back(piters);
+ CommandParameter psubsample("subsample", "String", "", "", "", "", "",false,false); parameters.push_back(psubsample);
+ CommandParameter pcutoff("cutoff", "Number", "", "10", "", "", "",false,false); parameters.push_back(pcutoff);
+ CommandParameter pprecision("precision", "Number", "", "100", "", "", "",false,false); parameters.push_back(pprecision);
+ CommandParameter plabel("label", "String", "", "", "", "", "",false,false); parameters.push_back(plabel);
+ CommandParameter pgroups("groups", "String", "", "", "", "", "",false,false); parameters.push_back(pgroups);
+ CommandParameter pcalc("calc", "Multiple", "sharedsobs-sharedchao-sharedace-jabund-sorabund-jclass-sorclass-jest-sorest-thetayc-thetan-kstest-sharednseqs-ochiai-anderberg-kulczynski-kulczynskicody-lennon-morisitahorn-braycurtis-whittaker-odum-canberra-structeuclidean-structchord-hellinger-manhattan-structpearson-soergel-spearman-structkulczynski-speciesprofile-hamming-structchi2-gower-memchi2-memchord-memeuclidean-mempearson", "jclass-thetayc", "", "", "",true,false); parameters.push_back(pcalc);
+
+ CommandParameter pprocessors("processors", "Number", "", "1", "", "", "",false,false); parameters.push_back(pprocessors);
+//CommandParameter poutput("output", "Multiple", "lt-square", "lt", "", "", "",false,false); parameters.push_back(poutput);
+ 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, "TreeGroupCommand", "setParameters");
+ exit(1);
+ }
+}
+//**********************************************************************************************************************
+string TreeGroupCommand::getHelpString(){
+ try {
+ string helpString = "";
+ ValidCalculators validCalculator;
+ helpString += "The tree.shared command creates a .tre to represent the similiarity between groups or sequences.\n";
+ helpString += "The tree.shared command parameters are shared, groups, calc, phylip, column, name, cutoff, precision, processors, subsample, iters and label.\n";
+ helpString += "The groups parameter allows you to specify which of the groups in your groupfile you would like included used.\n";
+ helpString += "The group names are separated by dashes. The label allow you to select what distance levels you would like trees created for, and are also separated by dashes.\n";
+ helpString += "The phylip or column parameter are required if you do not provide a sharedfile, and only one may be used. If you use a column file the name filename is required. \n";
+ helpString += "If you do not provide a cutoff value 10.00 is assumed. If you do not provide a precision value then 100 is assumed.\n";
+ helpString += "The tree.shared command should be in the following format: tree.shared(groups=yourGroups, calc=yourCalcs, label=yourLabels).\n";
+ helpString += "The iters parameter allows you to choose the number of times you would like to run the subsample.\n";
+ helpString += "The subsample parameter allows you to enter the size pergroup of the sample or you can set subsample=T and mothur will use the size of your smallest group. The subsample parameter may only be used with a shared file.\n";
+ helpString += "Example tree.shared(groups=A-B-C, calc=jabund-sorabund).\n";
+ helpString += "The default value for groups is all the groups in your groupfile.\n";
+ helpString += "The default value for calc is jclass-thetayc.\n";
+ helpString += "The tree.shared command outputs a .tre file for each calculator you specify at each distance you choose.\n";
+ helpString += validCalculator.printCalc("treegroup");
+ helpString += "Or the tree.shared command can be in the following format: tree.shared(phylip=yourPhylipFile).\n";
+ helpString += "Example tree.shared(phylip=abrecovery.dist).\n";
+ helpString += "Note: No spaces between parameter labels (i.e. groups), '=' and parameters (i.e.yourGroups).\n";
+ return helpString;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "TreeGroupCommand", "getHelpString");
+ exit(1);
+ }
+}
+//**********************************************************************************************************************
+string TreeGroupCommand::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 == "tree") { outputFileName = "tre"; }
+ 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, "TreeGroupCommand", "getOutputFileNameTag");
+ exit(1);
+ }
+}
+//**********************************************************************************************************************
+TreeGroupCommand::TreeGroupCommand(){
+ try {
+ abort = true; calledHelp = true;
+ setParameters();
+ //initialize outputTypes
+ vector<string> tempOutNames;
+ outputTypes["tree"] = tempOutNames;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "TreeGroupCommand", "TreeGroupCommand");
+ exit(1);
+ }
+}
//**********************************************************************************************************************
TreeGroupCommand::TreeGroupCommand(string option) {
try {
- globaldata = GlobalData::getInstance();
- abort = false;
+ abort = false; calledHelp = false;
allLines = 1;
- labels.clear();
- Groups.clear();
- Estimators.clear();
//allow user to run help
- if(option == "help") { validCalculator = new ValidCalculators(); 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 Array[] = {"label","calc","groups", "phylip", "column", "name", "precision","cutoff","outputdir","inputdir"};
- vector<string> myArray (Array, Array+(sizeof(Array)/sizeof(string)));
+ vector<string> myArray = setParameters();
OptionParser parser(option);
map<string, string> parameters = parser. getParameters();
if (validParameter.isValidParameter(it->first, myArray, it->second) != true) { abort = true; }
}
+ //initialize outputTypes
+ vector<string> tempOutNames;
+ outputTypes["tree"] = tempOutNames;
+
//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 = ""; }
it = parameters.find("phylip");
//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["phylip"] = inputDir + it->second; }
}
it = parameters.find("column");
//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["column"] = inputDir + it->second; }
}
it = parameters.find("name");
//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["name"] = 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; }
+ }
}
- format = globaldata->getFormat();
-
- //required parameters
+ //check for required parameters
phylipfile = validParameter.validFile(parameters, "phylip", true);
- if (phylipfile == "not open") { abort = true; }
+ if (phylipfile == "not open") { phylipfile = ""; abort = true; }
else if (phylipfile == "not found") { phylipfile = ""; }
- else { format = "phylip"; globaldata->setPhylipFile(phylipfile); }
+ else { inputfile = phylipfile; format = "phylip"; m->setPhylipFile(phylipfile); }
columnfile = validParameter.validFile(parameters, "column", true);
- if (columnfile == "not open") { abort = true; }
+ if (columnfile == "not open") { columnfile = ""; abort = true; }
else if (columnfile == "not found") { columnfile = ""; }
- else { format = "column"; globaldata->setColumnFile(columnfile); }
+ else { inputfile = columnfile; format = "column"; m->setColumnFile(columnfile); }
+
+ sharedfile = validParameter.validFile(parameters, "shared", true);
+ if (sharedfile == "not open") { sharedfile = ""; abort = true; }
+ else if (sharedfile == "not found") { sharedfile = ""; }
+ else { inputfile = sharedfile; format = "sharedfile"; m->setSharedFile(sharedfile); }
namefile = validParameter.validFile(parameters, "name", true);
if (namefile == "not open") { abort = true; }
else if (namefile == "not found") { namefile = ""; }
- else { globaldata->setNameFile(namefile); }
+ else { m->setNameFile(namefile); }
+
+ countfile = validParameter.validFile(parameters, "count", true);
+ if (countfile == "not open") { abort = true; countfile = ""; }
+ else if (countfile == "not found") { countfile = ""; }
+ else { m->setCountTableFile(countfile); }
- //error checking on files
- if ((globaldata->getSharedFile() == "") && ((phylipfile == "") && (columnfile == ""))) { m->mothurOut("You must run the read.otu command or provide a distance file before running the tree.shared command."); m->mothurOutEndLine(); abort = true; }
+ if ((phylipfile == "") && (columnfile == "") && (sharedfile == "")) {
+ //is there are current file available for either of these?
+ //give priority to shared, then column, then phylip
+ sharedfile = m->getSharedFile();
+ if (sharedfile != "") { inputfile = sharedfile; format = "sharedfile"; m->mothurOut("Using " + sharedfile + " as input file for the shared parameter."); m->mothurOutEndLine(); }
+ else {
+ columnfile = m->getColumnFile();
+ if (columnfile != "") { inputfile = columnfile; format = "column"; m->mothurOut("Using " + columnfile + " as input file for the column parameter."); m->mothurOutEndLine(); }
+ else {
+ phylipfile = m->getPhylipFile();
+ if (phylipfile != "") { inputfile = phylipfile; format = "phylip"; m->mothurOut("Using " + phylipfile + " as input file for the phylip parameter."); m->mothurOutEndLine(); }
+ else {
+ m->mothurOut("No valid current files. You must provide a shared, phylip or column file."); m->mothurOutEndLine();
+ abort = true;
+ }
+ }
+ }
+ }
else if ((phylipfile != "") && (columnfile != "")) { m->mothurOut("When running the tree.shared command with a distance file you may not use both the column and the phylip parameters."); m->mothurOutEndLine(); abort = true; }
if (columnfile != "") {
- if (namefile == "") { m->mothurOut("You need to provide a namefile if you are going to use the column format."); m->mothurOutEndLine(); abort = true; }
+ if ((namefile == "") && (countfile == "")){
+ namefile = m->getNameFile();
+ if (namefile != "") { m->mothurOut("Using " + namefile + " as input file for the name parameter."); m->mothurOutEndLine(); }
+ else {
+ countfile = m->getCountTableFile();
+ if (countfile != "") { m->mothurOut("Using " + countfile + " as input file for the count parameter."); m->mothurOutEndLine(); }
+ else {
+ m->mothurOut("You need to provide a namefile or countfile if you are going to use the column format."); m->mothurOutEndLine();
+ abort = true;
+ }
+ }
+ }
}
+
//check for optional parameter and set defaults
// ...at some point should added some additional type checking...
label = validParameter.validFile(parameters, "label", false);
if (label == "not found") { label = ""; }
else {
- if(label != "all") { splitAtDash(label, labels); allLines = 0; }
+ if(label != "all") { m->splitAtDash(label, labels); allLines = 0; }
else { allLines = 1; }
}
- //if the user has not specified any labels use the ones from read.otu
- if(label == "") {
- allLines = globaldata->allLines;
- labels = globaldata->labels;
- }
-
groups = validParameter.validFile(parameters, "groups", false);
if (groups == "not found") { groups = ""; }
else {
- splitAtDash(groups, Groups);
- globaldata->Groups = Groups;
+ m->splitAtDash(groups, Groups);
+ m->setGroups(Groups);
}
calc = validParameter.validFile(parameters, "calc", false);
else {
if (calc == "default") { calc = "jclass-thetayc"; }
}
- splitAtDash(calc, Estimators);
+ m->splitAtDash(calc, Estimators);
+ if (m->inUsersGroups("citation", Estimators)) {
+ ValidCalculators validCalc; validCalc.printCitations(Estimators);
+ //remove citation from list of calcs
+ for (int i = 0; i < Estimators.size(); i++) { if (Estimators[i] == "citation") { Estimators.erase(Estimators.begin()+i); break; } }
+ }
string temp;
temp = validParameter.validFile(parameters, "precision", false); if (temp == "not found") { temp = "100"; }
- convert(temp, precision);
+ m->mothurConvert(temp, precision);
temp = validParameter.validFile(parameters, "cutoff", false); if (temp == "not found") { temp = "10"; }
- convert(temp, cutoff);
+ m->mothurConvert(temp, cutoff);
cutoff += (5 / (precision * 10.0));
+ temp = validParameter.validFile(parameters, "processors", false); if (temp == "not found"){ temp = m->getProcessors(); }
+ m->setProcessors(temp);
+ m->mothurConvert(temp, processors);
+
+ temp = validParameter.validFile(parameters, "iters", false); if (temp == "not found") { temp = "1000"; }
+ m->mothurConvert(temp, iters);
+
+ temp = validParameter.validFile(parameters, "subsample", false); if (temp == "not found") { temp = "F"; }
+ if (m->isNumeric1(temp)) { m->mothurConvert(temp, subsampleSize); subsample = true; }
+ else {
+ if (m->isTrue(temp)) { subsample = true; subsampleSize = -1; } //we will set it to smallest group later
+ else { subsample = false; }
+ }
+
+ if (subsample == false) { iters = 1; }
+
+ if (subsample && (format != "sharedfile")) { m->mothurOut("[ERROR]: the subsample parameter can only be used with a shared file.\n"); 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 = "";
- outputDir += hasPath(globaldata->inputFileName); //if user entered a file with a path then preserve it
+ outputDir += m->hasPath(inputfile); //if user entered a file with a path then preserve it
}
-
-
- if (abort == false) {
-
- validCalculator = new ValidCalculators();
-
- if (format == "sharedfile") {
- int i;
- for (i=0; i<Estimators.size(); i++) {
- if (validCalculator->isValidCalculator("treegroup", Estimators[i]) == true) {
- if (Estimators[i] == "jabund") {
- treeCalculators.push_back(new JAbund());
- }else if (Estimators[i] == "sorabund") {
- treeCalculators.push_back(new SorAbund());
- }else if (Estimators[i] == "jclass") {
- treeCalculators.push_back(new Jclass());
- }else if (Estimators[i] == "sorclass") {
- treeCalculators.push_back(new SorClass());
- }else if (Estimators[i] == "jest") {
- treeCalculators.push_back(new Jest());
- }else if (Estimators[i] == "sorest") {
- treeCalculators.push_back(new SorEst());
- }else if (Estimators[i] == "thetayc") {
- treeCalculators.push_back(new ThetaYC());
- }else if (Estimators[i] == "thetan") {
- treeCalculators.push_back(new ThetaN());
- }else if (Estimators[i] == "morisitahorn") {
- treeCalculators.push_back(new MorHorn());
- }else if (Estimators[i] == "braycurtis") {
- treeCalculators.push_back(new BrayCurtis());
- }
- }
- }
- }
- }
}
}
exit(1);
}
}
-
-//**********************************************************************************************************************
-
-void TreeGroupCommand::help(){
- try {
- m->mothurOut("The tree.shared command creates a .tre to represent the similiarity between groups or sequences.\n");
- m->mothurOut("The tree.shared command can only be executed after a successful read.otu command or by providing a distance file.\n");
- m->mothurOut("The tree.shared command parameters are groups, calc, phylip, column, name, cutoff, precision and label.\n");
- m->mothurOut("The groups parameter allows you to specify which of the groups in your groupfile you would like included used.\n");
- m->mothurOut("The group names are separated by dashes. The label allow you to select what distance levels you would like trees created for, and are also separated by dashes.\n");
- m->mothurOut("The phylip or column parameter are required if you do not run the read.otu command first, and only one may be used. If you use a column file the name filename is required. \n");
- m->mothurOut("If you do not provide a cutoff value 10.00 is assumed. If you do not provide a precision value then 100 is assumed.\n");
- m->mothurOut("The tree.shared command should be in the following format: tree.shared(groups=yourGroups, calc=yourCalcs, label=yourLabels).\n");
- m->mothurOut("Example tree.shared(groups=A-B-C, calc=jabund-sorabund).\n");
- m->mothurOut("The default value for groups is all the groups in your groupfile.\n");
- m->mothurOut("The default value for calc is jclass-thetayc.\n");
- m->mothurOut("The tree.shared command outputs a .tre file for each calculator you specify at each distance you choose.\n");
- validCalculator->printCalc("treegroup", cout);
- m->mothurOut("Or the tree.shared command can be in the following format: tree.shared(phylip=yourPhylipFile).\n");
- m->mothurOut("Example tree.shared(phylip=abrecovery.dist).\n");
- m->mothurOut("Note: No spaces between parameter labels (i.e. groups), '=' and parameters (i.e.yourGroups).\n\n");
- }
- catch(exception& e) {
- m->errorOut(e, "TreeGroupCommand", "help");
- exit(1);
- }
-}
-
-
//**********************************************************************************************************************
TreeGroupCommand::~TreeGroupCommand(){
if (abort == false) {
-
- if (format == "sharedfile") { delete read; delete input; globaldata->ginput = NULL; }
- else { delete readMatrix; delete matrix; delete list; }
- delete tmap; globaldata->gTreemap = NULL;
- delete validCalculator;
+ if (format == "sharedfile") { delete input; }
+ else { delete list; }
+ delete ct;
}
}
int TreeGroupCommand::execute(){
try {
- if (abort == true) { return 0; }
+ if (abort == true) { if (calledHelp) { return 0; } return 2; }
if (format == "sharedfile") {
+
+ ValidCalculators validCalculator;
+
+ for (int i=0; i<Estimators.size(); i++) {
+ if (validCalculator.isValidCalculator("treegroup", Estimators[i]) == true) {
+ if (Estimators[i] == "sharedsobs") {
+ treeCalculators.push_back(new SharedSobsCS());
+ }else if (Estimators[i] == "sharedchao") {
+ treeCalculators.push_back(new SharedChao1());
+ }else if (Estimators[i] == "sharedace") {
+ treeCalculators.push_back(new SharedAce());
+ }else if (Estimators[i] == "jabund") {
+ treeCalculators.push_back(new JAbund());
+ }else if (Estimators[i] == "sorabund") {
+ treeCalculators.push_back(new SorAbund());
+ }else if (Estimators[i] == "jclass") {
+ treeCalculators.push_back(new Jclass());
+ }else if (Estimators[i] == "sorclass") {
+ treeCalculators.push_back(new SorClass());
+ }else if (Estimators[i] == "jest") {
+ treeCalculators.push_back(new Jest());
+ }else if (Estimators[i] == "sorest") {
+ treeCalculators.push_back(new SorEst());
+ }else if (Estimators[i] == "thetayc") {
+ treeCalculators.push_back(new ThetaYC());
+ }else if (Estimators[i] == "thetan") {
+ treeCalculators.push_back(new ThetaN());
+ }else if (Estimators[i] == "kstest") {
+ treeCalculators.push_back(new KSTest());
+ }else if (Estimators[i] == "sharednseqs") {
+ treeCalculators.push_back(new SharedNSeqs());
+ }else if (Estimators[i] == "ochiai") {
+ treeCalculators.push_back(new Ochiai());
+ }else if (Estimators[i] == "anderberg") {
+ treeCalculators.push_back(new Anderberg());
+ }else if (Estimators[i] == "kulczynski") {
+ treeCalculators.push_back(new Kulczynski());
+ }else if (Estimators[i] == "kulczynskicody") {
+ treeCalculators.push_back(new KulczynskiCody());
+ }else if (Estimators[i] == "lennon") {
+ treeCalculators.push_back(new Lennon());
+ }else if (Estimators[i] == "morisitahorn") {
+ treeCalculators.push_back(new MorHorn());
+ }else if (Estimators[i] == "braycurtis") {
+ treeCalculators.push_back(new BrayCurtis());
+ }else if (Estimators[i] == "whittaker") {
+ treeCalculators.push_back(new Whittaker());
+ }else if (Estimators[i] == "odum") {
+ treeCalculators.push_back(new Odum());
+ }else if (Estimators[i] == "canberra") {
+ treeCalculators.push_back(new Canberra());
+ }else if (Estimators[i] == "structeuclidean") {
+ treeCalculators.push_back(new StructEuclidean());
+ }else if (Estimators[i] == "structchord") {
+ treeCalculators.push_back(new StructChord());
+ }else if (Estimators[i] == "hellinger") {
+ treeCalculators.push_back(new Hellinger());
+ }else if (Estimators[i] == "manhattan") {
+ treeCalculators.push_back(new Manhattan());
+ }else if (Estimators[i] == "structpearson") {
+ treeCalculators.push_back(new StructPearson());
+ }else if (Estimators[i] == "soergel") {
+ treeCalculators.push_back(new Soergel());
+ }else if (Estimators[i] == "spearman") {
+ treeCalculators.push_back(new Spearman());
+ }else if (Estimators[i] == "structkulczynski") {
+ treeCalculators.push_back(new StructKulczynski());
+ }else if (Estimators[i] == "speciesprofile") {
+ treeCalculators.push_back(new SpeciesProfile());
+ }else if (Estimators[i] == "hamming") {
+ treeCalculators.push_back(new Hamming());
+ }else if (Estimators[i] == "structchi2") {
+ treeCalculators.push_back(new StructChi2());
+ }else if (Estimators[i] == "gower") {
+ treeCalculators.push_back(new Gower());
+ }else if (Estimators[i] == "memchi2") {
+ treeCalculators.push_back(new MemChi2());
+ }else if (Estimators[i] == "memchord") {
+ treeCalculators.push_back(new MemChord());
+ }else if (Estimators[i] == "memeuclidean") {
+ treeCalculators.push_back(new MemEuclidean());
+ }else if (Estimators[i] == "mempearson") {
+ treeCalculators.push_back(new MemPearson());
+ }
+ }
+ }
+
//if the users entered no valid calculators don't execute command
if (treeCalculators.size() == 0) { m->mothurOut("You have given no valid calculators."); m->mothurOutEndLine(); return 0; }
-
- //you have groups
- read = new ReadOTUFile(globaldata->inputFileName);
- read->read(&*globaldata);
- input = globaldata->ginput;
+ input = new InputData(sharedfile, "sharedfile");
lookup = input->getSharedRAbundVectors();
lastLabel = lookup[0]->getLabel();
if (lookup.size() < 2) { m->mothurOut("You have not provided enough valid groups. I cannot run the command."); m->mothurOutEndLine(); return 0; }
//used in tree constructor
- globaldata->runParse = false;
+ m->runParse = false;
+
+ //create treemap class from groupmap for tree class to use
+ ct = new CountTable();
+ set<string> nameMap;
+ map<string, string> groupMap;
+ set<string> gps;
+ for (int i = 0; i < m->getAllGroups().size(); i++) {
+ nameMap.insert(m->getAllGroups()[i]);
+ gps.insert(m->getAllGroups()[i]);
+ groupMap[m->getAllGroups()[i]] = m->getAllGroups()[i];
+ }
+ ct->createTable(nameMap, groupMap, gps);
+
+ //clear globaldatas old tree names if any
+ m->Treenames.clear();
+
+ //fills globaldatas tree names
+ m->Treenames = m->getGroups();
+
+ if (m->control_pressed) { return 0; }
//create tree file
makeSimsShared();
+
+ if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } return 0; }
}else{
//read in dist file
- filename = globaldata->inputFileName;
-
+ filename = inputfile;
+
+ ReadMatrix* readMatrix;
if (format == "column") { readMatrix = new ReadColumnMatrix(filename); }
else if (format == "phylip") { readMatrix = new ReadPhylipMatrix(filename); }
readMatrix->setCutoff(cutoff);
- if(namefile != ""){
- nameMap = new NameAssignment(namefile);
- nameMap->readMap();
- }
- else{
- nameMap = NULL;
- }
-
- readMatrix->read(nameMap);
+ ct = NULL;
+ if(namefile != ""){
+ nameMap = new NameAssignment(namefile);
+ nameMap->readMap();
+ readMatrix->read(nameMap);
+ }else if (countfile != "") {
+ ct = new CountTable();
+ ct->readTable(countfile);
+ readMatrix->read(ct);
+ }
+
list = readMatrix->getListVector();
- matrix = readMatrix->getMatrix();
+ SparseDistanceMatrix* dMatrix = readMatrix->getDMatrix();
//make treemap
- tmap = new TreeMap();
- tmap->makeSim(list);
- globaldata->gTreemap = tmap;
+ if (ct != NULL) { delete ct; }
+ ct = new CountTable();
+ set<string> nameMap;
+ map<string, string> groupMap;
+ set<string> gps;
+ for (int i = 0; i < list->getNumBins(); i++) {
+ string bin = list->get(i);
+ nameMap.insert(bin);
+ gps.insert(bin);
+ groupMap[bin] = bin;
+ }
+ ct->createTable(nameMap, groupMap, gps);
- globaldata->Groups = tmap->namesOfGroups;
+ vector<string> namesGroups = ct->getNamesOfGroups();
+ m->setGroups(namesGroups);
//clear globaldatas old tree names if any
- globaldata->Treenames.clear();
-
+ m->Treenames.clear();
+
//fills globaldatas tree names
- globaldata->Treenames = globaldata->Groups;
+ m->Treenames = m->getGroups();
//used in tree constructor
- globaldata->runParse = false;
+ m->runParse = false;
+
+ if (m->control_pressed) { return 0; }
- makeSimsDist();
+ vector< vector<double> > matrix = makeSimsDist(dMatrix);
+ delete readMatrix;
+ delete dMatrix;
+
+ if (m->control_pressed) { return 0; }
//create a new filename
- outputFile = outputDir + getRootName(getSimpleName(globaldata->inputFileName)) + "tre";
- outputNames.push_back(outputFile);
+ string outputFile = outputDir + m->getRootName(m->getSimpleName(inputfile)) + getOutputFileNameTag("tree");
+ outputNames.push_back(outputFile); outputTypes["tree"].push_back(outputFile);
- createTree();
+ Tree* newTree = createTree(matrix);
+
+ if (newTree != NULL) { writeTree(outputFile, newTree); delete newTree; }
+
+ if (m->control_pressed) { return 0; }
+
m->mothurOut("Tree complete. "); m->mothurOutEndLine();
+
}
//reset groups parameter
- globaldata->Groups.clear();
+ m->clearGroups();
+
+ //set tree file as new current treefile
+ string current = "";
+ itTypes = outputTypes.find("tree");
+ if (itTypes != outputTypes.end()) {
+ if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setTreeFile(current); }
+ }
m->mothurOutEndLine();
m->mothurOut("Output File Names: "); m->mothurOutEndLine();
}
//**********************************************************************************************************************
-void TreeGroupCommand::createTree(){
+Tree* TreeGroupCommand::createTree(vector< vector<double> >& simMatrix){
try {
//create tree
- t = new Tree();
-
- //do merges and create tree structure by setting parents and children
- //there are numGroups - 1 merges to do
- for (int i = 0; i < (numGroups - 1); i++) {
- float largest = -1000.0;
-
- int row, column;
- //find largest value in sims matrix by searching lower triangle
- for (int j = 1; j < simMatrix.size(); j++) {
- for (int k = 0; k < j; k++) {
- if (simMatrix[j][k] > largest) { largest = simMatrix[j][k]; row = j; column = k; }
- }
- }
-
- //set non-leaf node info and update leaves to know their parents
- //non-leaf
- t->tree[numGroups + i].setChildren(index[row], index[column]);
-
- //parents
- t->tree[index[row]].setParent(numGroups + i);
- t->tree[index[column]].setParent(numGroups + i);
-
- //blength = distance / 2;
- float blength = ((1.0 - largest) / 2);
-
- //branchlengths
- t->tree[index[row]].setBranchLength(blength - t->tree[index[row]].getLengthToLeaves());
- t->tree[index[column]].setBranchLength(blength - t->tree[index[column]].getLengthToLeaves());
-
- //set your length to leaves to your childs length plus branchlength
- t->tree[numGroups + i].setLengthToLeaves(t->tree[index[row]].getLengthToLeaves() + t->tree[index[row]].getBranchLength());
-
-
- //update index
- index[row] = numGroups+i;
- index[column] = numGroups+i;
-
- //remove highest value that caused the merge.
- simMatrix[row][column] = -1000.0;
- simMatrix[column][row] = -1000.0;
-
- //merge values in simsMatrix
- for (int n = 0; n < simMatrix.size(); n++) {
- //row becomes merge of 2 groups
- simMatrix[row][n] = (simMatrix[row][n] + simMatrix[column][n]) / 2;
- simMatrix[n][row] = simMatrix[row][n];
- //delete column
- simMatrix[column][n] = -1000.0;
- simMatrix[n][column] = -1000.0;
- }
- }
-
- //adjust tree to make sure root to tip length is .5
- int root = t->findRoot();
- t->tree[root].setBranchLength((0.5 - t->tree[root].getLengthToLeaves()));
+ t = new Tree(ct, simMatrix);
+
+ if (m->control_pressed) { delete t; t = NULL; return t; }
- //assemble tree
+ //assemble tree
t->assembleTree();
-
- //print newick file
- t->createNewickFile(outputFile);
-
- //delete tree
- delete t;
-
+
+ return t;
}
catch(exception& e) {
m->errorOut(e, "TreeGroupCommand", "createTree");
}
}
/***********************************************************/
-void TreeGroupCommand::printSims(ostream& out) {
+int TreeGroupCommand::writeTree(string out, Tree* T) {
try {
- //output column headers
- //out << '\t';
- //for (int i = 0; i < lookup.size(); i++) { out << lookup[i]->getGroup() << '\t'; }
- //out << endl;
+ //print newick file
+ t->createNewickFile(out);
+ if (m->control_pressed) { m->mothurRemove(out); outputNames.pop_back(); return 1; }
+
+ return 0;
+
+ }
+ catch(exception& e) {
+ m->errorOut(e, "TreeGroupCommand", "printSims");
+ exit(1);
+ }
+}
+
+/***********************************************************/
+void TreeGroupCommand::printSims(ostream& out, vector< vector<double> >& simMatrix) {
+ try {
- for (int m = 0; m < simMatrix.size(); m++) {
+ for (int m = 0; m < simMatrix.size(); m++) {
//out << lookup[m]->getGroup() << '\t';
for (int n = 0; n < simMatrix.size(); n++) {
out << simMatrix[m][n] << '\t';
}
}
/***********************************************************/
-void TreeGroupCommand::makeSimsDist() {
+vector< vector<double> > TreeGroupCommand::makeSimsDist(SparseDistanceMatrix* matrix) {
try {
numGroups = list->size();
- //initialize index
- index.clear();
- for (int g = 0; g < numGroups; g++) { index[g] = g; }
-
//initialize simMatrix
- simMatrix.clear();
+ vector< vector<double> > simMatrix;
simMatrix.resize(numGroups);
- for (int m = 0; m < simMatrix.size(); m++) {
+ for (int k = 0; k < simMatrix.size(); k++) {
for (int j = 0; j < simMatrix.size(); j++) {
- simMatrix[m].push_back(0.0);
+ simMatrix[k].push_back(0.0);
}
}
//go through sparse matrix and fill sims
//go through each cell in the sparsematrix
- for(MatData currentCell = matrix->begin(); currentCell != matrix->end(); currentCell++){
- //similairity = -(distance-1)
- simMatrix[currentCell->row][currentCell->column] = -(currentCell->dist -1.0);
- simMatrix[currentCell->column][currentCell->row] = -(currentCell->dist -1.0);
+ for (int i = 0; i < matrix->seqVec.size(); i++) {
+ for (int j = 0; j < matrix->seqVec[i].size(); j++) {
+
+ //already checked everyone else in row
+ if (i < matrix->seqVec[i][j].index) {
+ simMatrix[i][matrix->seqVec[i][j].index] = -(matrix->seqVec[i][j].dist -1.0);
+ simMatrix[matrix->seqVec[i][j].index][i] = -(matrix->seqVec[i][j].dist -1.0);
+
+ if (m->control_pressed) { return simMatrix; }
+ }
+ }
}
-
+ return simMatrix;
}
catch(exception& e) {
m->errorOut(e, "TreeGroupCommand", "makeSimsDist");
}
/***********************************************************/
-void TreeGroupCommand::makeSimsShared() {
+int TreeGroupCommand::makeSimsShared() {
try {
-
- //clear globaldatas old tree names if any
- globaldata->Treenames.clear();
-
- //fills globaldatas tree names
- globaldata->Treenames = globaldata->Groups;
-
- //create treemap class from groupmap for tree class to use
- tmap = new TreeMap();
- tmap->makeSim(globaldata->gGroupmap);
- globaldata->gTreemap = tmap;
-
+
+ if (subsample) {
+ if (subsampleSize == -1) { //user has not set size, set size = smallest samples size
+ subsampleSize = lookup[0]->getNumSeqs();
+ for (int i = 1; i < lookup.size(); i++) {
+ int thisSize = lookup[i]->getNumSeqs();
+
+ if (thisSize < subsampleSize) { subsampleSize = thisSize; }
+ }
+ }else {
+ m->clearGroups();
+ Groups.clear();
+ vector<SharedRAbundVector*> temp;
+ for (int i = 0; i < lookup.size(); i++) {
+ if (lookup[i]->getNumSeqs() < subsampleSize) {
+ m->mothurOut(lookup[i]->getGroup() + " contains " + toString(lookup[i]->getNumSeqs()) + ". Eliminating."); m->mothurOutEndLine();
+ delete lookup[i];
+ }else {
+ Groups.push_back(lookup[i]->getGroup());
+ temp.push_back(lookup[i]);
+ }
+ }
+ lookup = temp;
+ m->setGroups(Groups);
+ }
+
+ if (lookup.size() < 2) { m->mothurOut("You have not provided enough valid groups. I cannot run the command."); m->mothurOutEndLine(); m->control_pressed = true; return 0; }
+ }
+
+ numGroups = lookup.size();
+ lines.resize(processors);
+ for (int i = 0; i < processors; i++) {
+ lines[i].start = int (sqrt(float(i)/float(processors)) * numGroups);
+ lines[i].end = int (sqrt(float(i+1)/float(processors)) * numGroups);
+ }
+
set<string> processedLabels;
set<string> userLabels = labels;
//as long as you are not at the end of the file or done wih the lines you want
while((lookup[0] != NULL) && ((allLines == 1) || (userLabels.size() != 0))) {
+ if (m->control_pressed) { for (int i = 0; i < lookup.size(); i++) { delete lookup[i]; } for(int i = 0 ; i < treeCalculators.size(); i++) { delete treeCalculators[i]; } return 1; }
if(allLines == 1 || labels.count(lookup[0]->getLabel()) == 1){
m->mothurOut(lookup[0]->getLabel()); m->mothurOutEndLine();
userLabels.erase(lookup[0]->getLabel());
}
- if ((anyLabelsToProcess(lookup[0]->getLabel(), userLabels, "") == true) && (processedLabels.count(lastLabel) != 1)) {
+ if ((m->anyLabelsToProcess(lookup[0]->getLabel(), userLabels, "") == true) && (processedLabels.count(lastLabel) != 1)) {
string saveLabel = lookup[0]->getLabel();
for (int i = 0; i < lookup.size(); i++) { delete lookup[i]; }
lookup = input->getSharedRAbundVectors();
}
+ if (m->control_pressed) { for (int i = 0; i < lookup.size(); i++) { delete lookup[i]; } for(int i = 0 ; i < treeCalculators.size(); i++) { delete treeCalculators[i]; } return 1; }
+
//output error messages about any remaining user labels
set<string>::iterator it;
bool needToRun = false;
}
for(int i = 0 ; i < treeCalculators.size(); i++) { delete treeCalculators[i]; }
+
+ return 0;
}
catch(exception& e) {
m->errorOut(e, "TreeGroupCommand", "makeSimsShared");
}
/***********************************************************/
-void TreeGroupCommand::process(vector<SharedRAbundVector*> thisLookup) {
+int TreeGroupCommand::process(vector<SharedRAbundVector*> thisLookup) {
try{
- EstOutput data;
- vector<SharedRAbundVector*> subset;
- numGroups = thisLookup.size();
-
- //for each calculator
- for(int i = 0 ; i < treeCalculators.size(); i++) {
- //initialize simMatrix
- simMatrix.clear();
- simMatrix.resize(numGroups);
- for (int m = 0; m < simMatrix.size(); m++) {
- for (int j = 0; j < simMatrix.size(); j++) {
- simMatrix[m].push_back(0.0);
- }
- }
+ vector< vector< vector<seqDist> > > calcDistsTotals; //each iter, one for each calc, then each groupCombos dists. this will be used to make .dist files
+ vector< vector<seqDist> > calcDists; calcDists.resize(treeCalculators.size());
+
+ for (int thisIter = 0; thisIter < iters; thisIter++) {
+
+ vector<SharedRAbundVector*> thisItersLookup = thisLookup;
+
+ if (subsample) {
+ SubSample sample;
+ vector<string> tempLabels; //dont need since we arent printing the sampled sharedRabunds
+
+ //make copy of lookup so we don't get access violations
+ vector<SharedRAbundVector*> newLookup;
+ for (int k = 0; k < thisItersLookup.size(); k++) {
+ SharedRAbundVector* temp = new SharedRAbundVector();
+ temp->setLabel(thisItersLookup[k]->getLabel());
+ temp->setGroup(thisItersLookup[k]->getGroup());
+ newLookup.push_back(temp);
+ }
+
+ //for each bin
+ for (int k = 0; k < thisItersLookup[0]->getNumBins(); k++) {
+ if (m->control_pressed) { for (int j = 0; j < newLookup.size(); j++) { delete newLookup[j]; } return 0; }
+ for (int j = 0; j < thisItersLookup.size(); j++) { newLookup[j]->push_back(thisItersLookup[j]->getAbundance(k), thisItersLookup[j]->getGroup()); }
+ }
+
+ tempLabels = sample.getSample(newLookup, subsampleSize);
+ thisItersLookup = newLookup;
+ }
+
+ if(processors == 1){
+ driver(thisItersLookup, 0, numGroups, calcDists);
+ }else{
+ int process = 1;
+ vector<int> processIDS;
+
+#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) || (__linux__) || (__unix__) || (__unix)
+ //loop through and create all the processes you want
+ while (process != processors) {
+ int pid = fork();
+
+ if (pid > 0) {
+ processIDS.push_back(pid);
+ process++;
+ }else if (pid == 0){
+
+ driver(thisItersLookup, lines[process].start, lines[process].end, calcDists);
+
+ string tempdistFileName = m->getRootName(m->getSimpleName(sharedfile)) + toString(getpid()) + ".dist";
+ ofstream outtemp;
+ m->openOutputFile(tempdistFileName, outtemp);
+
+ for (int i = 0; i < calcDists.size(); i++) {
+ outtemp << calcDists[i].size() << endl;
+
+ for (int j = 0; j < calcDists[i].size(); j++) {
+ outtemp << calcDists[i][j].seq1 << '\t' << calcDists[i][j].seq2 << '\t' << calcDists[i][j].dist << endl;
+ }
+ }
+ outtemp.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);
+ }
+ }
+
+ //parent do your part
+ driver(thisItersLookup, lines[0].start, lines[0].end, calcDists);
+
+ //force parent to wait until all the processes are done
+ for (int i = 0; i < processIDS.size(); i++) {
+ int temp = processIDS[i];
+ wait(&temp);
+ }
+
+ for (int i = 0; i < processIDS.size(); i++) {
+ string tempdistFileName = m->getRootName(m->getSimpleName(sharedfile)) + toString(processIDS[i]) + ".dist";
+ ifstream intemp;
+ m->openInputFile(tempdistFileName, intemp);
+
+ for (int k = 0; k < calcDists.size(); k++) {
+ int size = 0;
+ intemp >> size; m->gobble(intemp);
+
+ for (int j = 0; j < size; j++) {
+ int seq1 = 0;
+ int seq2 = 0;
+ float dist = 1.0;
+
+ intemp >> seq1 >> seq2 >> dist; m->gobble(intemp);
+
+ seqDist tempDist(seq1, seq2, dist);
+ calcDists[k].push_back(tempDist);
+ }
+ }
+ intemp.close();
+ m->mothurRemove(tempdistFileName);
+ }
+#else
+ //////////////////////////////////////////////////////////////////////////////////////////////////////
+ //Windows version shared memory, so be careful when passing variables through the treeSharedData struct.
+ //Above fork() will clone, so memory is separate, but that's not the case with windows,
+ //Taking advantage of shared memory to pass results vectors.
+ //////////////////////////////////////////////////////////////////////////////////////////////////////
+
+ vector<treeSharedData*> pDataArray;
+ DWORD dwThreadIdArray[processors-1];
+ HANDLE hThreadArray[processors-1];
+
+ //Create processor worker threads.
+ for( int i=1; i<processors; i++ ){
+
+ //make copy of lookup so we don't get access violations
+ vector<SharedRAbundVector*> newLookup;
+ for (int k = 0; k < thisItersLookup.size(); k++) {
+ SharedRAbundVector* temp = new SharedRAbundVector();
+ temp->setLabel(thisItersLookup[k]->getLabel());
+ temp->setGroup(thisItersLookup[k]->getGroup());
+ newLookup.push_back(temp);
+ }
+
+ //for each bin
+ for (int k = 0; k < thisItersLookup[0]->getNumBins(); k++) {
+ if (m->control_pressed) { for (int j = 0; j < newLookup.size(); j++) { delete newLookup[j]; } return 0; }
+ for (int j = 0; j < thisItersLookup.size(); j++) { newLookup[j]->push_back(thisItersLookup[j]->getAbundance(k), thisItersLookup[j]->getGroup()); }
+ }
+
+ // Allocate memory for thread data.
+ treeSharedData* tempSum = new treeSharedData(m, lines[i].start, lines[i].end, Estimators, newLookup);
+ pDataArray.push_back(tempSum);
+ processIDS.push_back(i);
+
+ hThreadArray[i-1] = CreateThread(NULL, 0, MyTreeSharedThreadFunction, pDataArray[i-1], 0, &dwThreadIdArray[i-1]);
+ }
+
+ //parent do your part
+ driver(thisItersLookup, lines[0].start, lines[0].end, calcDists);
+
+ //Wait until all threads have terminated.
+ WaitForMultipleObjects(processors-1, hThreadArray, TRUE, INFINITE);
+
+ //Close all thread handles and free memory allocations.
+ for(int i=0; i < pDataArray.size(); i++){
+ for (int j = 0; j < pDataArray[i]->thisLookup.size(); j++) { delete pDataArray[i]->thisLookup[j]; }
+
+ for (int k = 0; k < calcDists.size(); k++) {
+ int size = pDataArray[i]->calcDists[k].size();
+ for (int j = 0; j < size; j++) { calcDists[k].push_back(pDataArray[i]->calcDists[k][j]); }
+ }
+
+ CloseHandle(hThreadArray[i]);
+ delete pDataArray[i];
+ }
+
+#endif
+ }
+
+ calcDistsTotals.push_back(calcDists);
+
+ if (subsample) {
+
+ //clean up memory
+ for (int i = 0; i < thisItersLookup.size(); i++) { delete thisItersLookup[i]; }
+ thisItersLookup.clear();
+ for (int i = 0; i < calcDists.size(); i++) { calcDists[i].clear(); }
+ }
+ }
- //initialize index
- index.clear();
- for (int g = 0; g < numGroups; g++) { index[g] = g; }
+ if (iters != 1) {
+ //we need to find the average distance and standard deviation for each groups distance
+
+ vector< vector<seqDist> > calcAverages; calcAverages.resize(treeCalculators.size());
+ for (int i = 0; i < calcAverages.size(); i++) { //initialize sums to zero.
+ calcAverages[i].resize(calcDistsTotals[0][i].size());
+
+ for (int j = 0; j < calcAverages[i].size(); j++) {
+ calcAverages[i][j].seq1 = calcDists[i][j].seq1;
+ calcAverages[i][j].seq2 = calcDists[i][j].seq2;
+ calcAverages[i][j].dist = 0.0;
+ }
+ }
+
+ for (int thisIter = 0; thisIter < iters; thisIter++) { //sum all groups dists for each calculator
+ for (int i = 0; i < calcAverages.size(); i++) { //initialize sums to zero.
+ for (int j = 0; j < calcAverages[i].size(); j++) {
+ calcAverages[i][j].dist += calcDistsTotals[thisIter][i][j].dist;
+ }
+ }
+ }
+
+ for (int i = 0; i < calcAverages.size(); i++) { //finds average.
+ for (int j = 0; j < calcAverages[i].size(); j++) {
+ calcAverages[i][j].dist /= (float) iters;
+ }
+ }
+
+ //create average tree for each calc
+ for (int i = 0; i < calcDists.size(); i++) {
+ vector< vector<double> > matrix; //square matrix to represent the distance
+ matrix.resize(thisLookup.size());
+ for (int k = 0; k < thisLookup.size(); k++) { matrix[k].resize(thisLookup.size(), 0.0); }
+
+ for (int j = 0; j < calcAverages[i].size(); j++) {
+ int row = calcAverages[i][j].seq1;
+ int column = calcAverages[i][j].seq2;
+ float dist = calcAverages[i][j].dist;
+
+ matrix[row][column] = dist;
+ matrix[column][row] = dist;
+ }
+
+ //create a new filename
+ string outputFile = outputDir + m->getRootName(m->getSimpleName(inputfile)) + treeCalculators[i]->getName() + "." + thisLookup[0]->getLabel() + ".ave." + getOutputFileNameTag("tree");
+ outputNames.push_back(outputFile); outputTypes["tree"].push_back(outputFile);
+
+ //creates tree from similarity matrix and write out file
+ Tree* newTree = createTree(matrix);
+ if (newTree != NULL) { writeTree(outputFile, newTree); }
+ }
+
+ //create all trees for each calc and find their consensus tree
+ for (int i = 0; i < calcDists.size(); i++) {
+ if (m->control_pressed) { break; }
+
+ //create a new filename
+ string outputFile = outputDir + m->getRootName(m->getSimpleName(inputfile)) + treeCalculators[i]->getName() + "." + thisLookup[0]->getLabel() + ".all." + getOutputFileNameTag("tree");
+ outputNames.push_back(outputFile); outputTypes["tree"].push_back(outputFile);
+
+ ofstream outAll;
+ m->openOutputFile(outputFile, outAll);
+
+ vector<Tree*> trees;
+ for (int myIter = 0; myIter < iters; myIter++) {
+
+ if(m->control_pressed) { break; }
+
+ //initialize matrix
+ vector< vector<double> > matrix; //square matrix to represent the distance
+ matrix.resize(thisLookup.size());
+ for (int k = 0; k < thisLookup.size(); k++) { matrix[k].resize(thisLookup.size(), 0.0); }
+
+ for (int j = 0; j < calcDistsTotals[myIter][i].size(); j++) {
+ int row = calcDistsTotals[myIter][i][j].seq1;
+ int column = calcDistsTotals[myIter][i][j].seq2;
+ double dist = calcDistsTotals[myIter][i][j].dist;
+
+ matrix[row][column] = dist;
+ matrix[column][row] = dist;
+ }
+
+ //creates tree from similarity matrix and write out file
+ Tree* newTree = createTree(matrix);
+ if (newTree != NULL) {
+ newTree->print(outAll);
+ trees.push_back(newTree);
+ }
+ }
+ outAll.close();
+ if (m->control_pressed) { for (int k = 0; k < trees.size(); k++) { delete trees[k]; } }
+
+ Consensus consensus;
+ //clear old tree names if any
+ m->Treenames.clear(); m->Treenames = m->getGroups(); //may have changed if subsample eliminated groups
+ Tree* conTree = consensus.getTree(trees);
+
+ //create a new filename
+ string conFile = outputDir + m->getRootName(m->getSimpleName(inputfile)) + treeCalculators[i]->getName() + "." + thisLookup[0]->getLabel() + ".cons." + getOutputFileNameTag("tree");
+ outputNames.push_back(conFile); outputTypes["tree"].push_back(conFile);
+ ofstream outTree;
+ m->openOutputFile(conFile, outTree);
+
+ if (conTree != NULL) { conTree->print(outTree, "boot"); delete conTree; }
+ }
+
+ }else {
+
+ for (int i = 0; i < calcDists.size(); i++) {
+ if (m->control_pressed) { break; }
+
+ //initialize matrix
+ vector< vector<double> > matrix; //square matrix to represent the distance
+ matrix.resize(thisLookup.size());
+ for (int k = 0; k < thisLookup.size(); k++) { matrix[k].resize(thisLookup.size(), 0.0); }
+
+ for (int j = 0; j < calcDists[i].size(); j++) {
+ int row = calcDists[i][j].seq1;
+ int column = calcDists[i][j].seq2;
+ double dist = calcDists[i][j].dist;
+
+ matrix[row][column] = dist;
+ matrix[column][row] = dist;
+ }
+
+ //create a new filename
+ string outputFile = outputDir + m->getRootName(m->getSimpleName(inputfile)) + treeCalculators[i]->getName() + "." + thisLookup[0]->getLabel() + "." + getOutputFileNameTag("tree");
+ outputNames.push_back(outputFile); outputTypes["tree"].push_back(outputFile);
+
+ //creates tree from similarity matrix and write out file
+ Tree* newTree = createTree(matrix);
+ if (newTree != NULL) { writeTree(outputFile, newTree); delete newTree; }
+ }
+ }
- //create a new filename
- outputFile = outputDir + getRootName(getSimpleName(globaldata->inputFileName)) + treeCalculators[i]->getName() + "." + thisLookup[0]->getLabel() + ".tre";
- outputNames.push_back(outputFile);
-
- for (int k = 0; k < thisLookup.size(); k++) {
- for (int l = k; l < thisLookup.size(); l++) {
- if (k != l) { //we dont need to similiarity of a groups to itself
- //get estimated similarity between 2 groups
-
- subset.clear(); //clear out old pair of sharedrabunds
- //add new pair of sharedrabunds
- subset.push_back(thisLookup[k]); subset.push_back(thisLookup[l]);
-
- data = treeCalculators[i]->getValues(subset); //saves the calculator outputs
- //save values in similarity matrix
- simMatrix[k][l] = data[0];
- simMatrix[l][k] = data[0];
+ return 0;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "TreeGroupCommand", "process");
+ exit(1);
+ }
+}
+/**************************************************************************************************/
+int TreeGroupCommand::driver(vector<SharedRAbundVector*> thisLookup, int start, int end, vector< vector<seqDist> >& calcDists) {
+ try {
+ vector<SharedRAbundVector*> subset;
+ for (int k = start; k < end; k++) { // pass cdd each set of groups to compare
+
+ for (int l = 0; l < k; l++) {
+
+ if (k != l) { //we dont need to similiarity of a groups to itself
+ subset.clear(); //clear out old pair of sharedrabunds
+ //add new pair of sharedrabunds
+ subset.push_back(thisLookup[k]); subset.push_back(thisLookup[l]);
+
+ for(int i=0;i<treeCalculators.size();i++) {
+
+ //if this calc needs all groups to calculate the pair load all groups
+ if (treeCalculators[i]->getNeedsAll()) {
+ //load subset with rest of lookup for those calcs that need everyone to calc for a pair
+ for (int w = 0; w < thisLookup.size(); w++) {
+ if ((w != k) && (w != l)) { subset.push_back(thisLookup[w]); }
}
}
+
+ vector<double> tempdata = treeCalculators[i]->getValues(subset); //saves the calculator outputs
+
+ if (m->control_pressed) { return 1; }
+
+ seqDist temp(l, k, -(tempdata[0]-1.0));
+ calcDists[i].push_back(temp);
}
-
- //creates tree from similarity matrix and write out file
- createTree();
}
-
+ }
+ }
+
+ return 0;
}
catch(exception& e) {
- m->errorOut(e, "TreeGroupCommand", "process");
+ m->errorOut(e, "TreeGroupCommand", "driver");
exit(1);
}
}