2 * treegroupscommand.cpp
5 * Created by Sarah Westcott on 4/8/09.
6 * Copyright 2009 Schloss Lab UMASS Amherst. All rights reserved.
10 #include "treegroupscommand.h"
11 #include "sharedsobscollectsummary.h"
12 #include "sharedchao1.h"
13 #include "sharedace.h"
14 #include "sharednseqs.h"
15 #include "sharedjabund.h"
16 #include "sharedsorabund.h"
17 #include "sharedjclass.h"
18 #include "sharedsorclass.h"
19 #include "sharedjest.h"
20 #include "sharedsorest.h"
21 #include "sharedthetayc.h"
22 #include "sharedthetan.h"
23 #include "sharedkstest.h"
24 #include "whittaker.h"
25 #include "sharedochiai.h"
26 #include "sharedanderbergs.h"
27 #include "sharedkulczynski.h"
28 #include "sharedkulczynskicody.h"
29 #include "sharedlennon.h"
30 #include "sharedmorisitahorn.h"
31 #include "sharedbraycurtis.h"
32 #include "sharedjackknife.h"
33 #include "whittaker.h"
36 #include "structeuclidean.h"
37 #include "structchord.h"
38 #include "hellinger.h"
39 #include "manhattan.h"
40 #include "structpearson.h"
43 #include "structkulczynski.h"
44 #include "structchi2.h"
45 #include "speciesprofile.h"
50 #include "memeuclidean.h"
51 #include "mempearson.h"
53 //**********************************************************************************************************************
54 vector<string> TreeGroupCommand::getValidParameters(){
56 string Array[] = {"label","calc","groups", "phylip", "column", "name", "precision","cutoff","outputdir","inputdir"};
57 vector<string> myArray (Array, Array+(sizeof(Array)/sizeof(string)));
61 m->errorOut(e, "TreeGroupCommand", "getValidParameters");
65 //**********************************************************************************************************************
66 TreeGroupCommand::TreeGroupCommand(){
68 abort = true; calledHelp = true;
69 globaldata = GlobalData::getInstance();
70 //initialize outputTypes
71 vector<string> tempOutNames;
72 outputTypes["tree"] = tempOutNames;
75 m->errorOut(e, "TreeGroupCommand", "TreeGroupCommand");
79 //**********************************************************************************************************************
80 vector<string> TreeGroupCommand::getRequiredParameters(){
82 vector<string> myArray;
86 m->errorOut(e, "TreeGroupCommand", "getRequiredParameters");
90 //**********************************************************************************************************************
91 vector<string> TreeGroupCommand::getRequiredFiles(){
93 string Array[] = {"phylip","column","shared","or"};
94 vector<string> myArray (Array, Array+(sizeof(Array)/sizeof(string)));
98 m->errorOut(e, "TreeGroupCommand", "getRequiredFiles");
102 //**********************************************************************************************************************
104 TreeGroupCommand::TreeGroupCommand(string option) {
106 globaldata = GlobalData::getInstance();
107 abort = false; calledHelp = false;
113 //allow user to run help
114 if(option == "help") { validCalculator = new ValidCalculators(); help(); abort = true; calledHelp = true; }
117 //valid paramters for this command
118 string Array[] = {"label","calc","groups", "phylip", "column", "name", "precision","cutoff","outputdir","inputdir"};
119 vector<string> myArray (Array, Array+(sizeof(Array)/sizeof(string)));
121 OptionParser parser(option);
122 map<string, string> parameters = parser. getParameters();
124 ValidParameters validParameter;
125 map<string, string>::iterator it;
127 //check to make sure all parameters are valid for command
128 for (it = parameters.begin(); it != parameters.end(); it++) {
129 if (validParameter.isValidParameter(it->first, myArray, it->second) != true) { abort = true; }
132 //initialize outputTypes
133 vector<string> tempOutNames;
134 outputTypes["tree"] = tempOutNames;
136 //if the user changes the input directory command factory will send this info to us in the output parameter
137 string inputDir = validParameter.validFile(parameters, "inputdir", false);
138 if (inputDir == "not found"){ inputDir = ""; }
141 it = parameters.find("phylip");
142 //user has given a template file
143 if(it != parameters.end()){
144 path = m->hasPath(it->second);
145 //if the user has not given a path then, add inputdir. else leave path alone.
146 if (path == "") { parameters["phylip"] = inputDir + it->second; }
149 it = parameters.find("column");
150 //user has given a template file
151 if(it != parameters.end()){
152 path = m->hasPath(it->second);
153 //if the user has not given a path then, add inputdir. else leave path alone.
154 if (path == "") { parameters["column"] = inputDir + it->second; }
157 it = parameters.find("name");
158 //user has given a template file
159 if(it != parameters.end()){
160 path = m->hasPath(it->second);
161 //if the user has not given a path then, add inputdir. else leave path alone.
162 if (path == "") { parameters["name"] = inputDir + it->second; }
166 format = globaldata->getFormat();
168 //required parameters
169 phylipfile = validParameter.validFile(parameters, "phylip", true);
170 if (phylipfile == "not open") { abort = true; }
171 else if (phylipfile == "not found") { phylipfile = ""; }
172 else { globaldata->newRead(); format = "phylip"; globaldata->setPhylipFile(phylipfile); }
174 columnfile = validParameter.validFile(parameters, "column", true);
175 if (columnfile == "not open") { abort = true; }
176 else if (columnfile == "not found") { columnfile = ""; }
177 else { globaldata->newRead(); format = "column"; globaldata->setColumnFile(columnfile); }
179 namefile = validParameter.validFile(parameters, "name", true);
180 if (namefile == "not open") { abort = true; }
181 else if (namefile == "not found") { namefile = ""; }
182 else { globaldata->setNameFile(namefile); }
184 //error checking on files
185 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; }
186 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; }
188 if (columnfile != "") {
189 if (namefile == "") { m->mothurOut("You need to provide a namefile if you are going to use the column format."); m->mothurOutEndLine(); abort = true; }
192 //check for optional parameter and set defaults
193 // ...at some point should added some additional type checking...
194 label = validParameter.validFile(parameters, "label", false);
195 if (label == "not found") { label = ""; }
197 if(label != "all") { m->splitAtDash(label, labels); allLines = 0; }
198 else { allLines = 1; }
201 //if the user has not specified any labels use the ones from read.otu
203 allLines = globaldata->allLines;
204 labels = globaldata->labels;
207 groups = validParameter.validFile(parameters, "groups", false);
208 if (groups == "not found") { groups = ""; }
210 m->splitAtDash(groups, Groups);
211 globaldata->Groups = Groups;
214 calc = validParameter.validFile(parameters, "calc", false);
215 if (calc == "not found") { calc = "jclass-thetayc"; }
217 if (calc == "default") { calc = "jclass-thetayc"; }
219 m->splitAtDash(calc, Estimators);
222 temp = validParameter.validFile(parameters, "precision", false); if (temp == "not found") { temp = "100"; }
223 convert(temp, precision);
225 temp = validParameter.validFile(parameters, "cutoff", false); if (temp == "not found") { temp = "10"; }
226 convert(temp, cutoff);
227 cutoff += (5 / (precision * 10.0));
229 //if the user changes the output directory command factory will send this info to us in the output parameter
230 outputDir = validParameter.validFile(parameters, "outputdir", false); if (outputDir == "not found"){
232 outputDir += m->hasPath(globaldata->inputFileName); //if user entered a file with a path then preserve it
236 if (abort == false) {
238 validCalculator = new ValidCalculators();
240 if (format == "sharedfile") {
242 for (i=0; i<Estimators.size(); i++) {
243 if (validCalculator->isValidCalculator("treegroup", Estimators[i]) == true) {
244 if (Estimators[i] == "sharedsobs") {
245 treeCalculators.push_back(new SharedSobsCS());
246 }else if (Estimators[i] == "sharedchao") {
247 treeCalculators.push_back(new SharedChao1());
248 }else if (Estimators[i] == "sharedace") {
249 treeCalculators.push_back(new SharedAce());
250 }else if (Estimators[i] == "jabund") {
251 treeCalculators.push_back(new JAbund());
252 }else if (Estimators[i] == "sorabund") {
253 treeCalculators.push_back(new SorAbund());
254 }else if (Estimators[i] == "jclass") {
255 treeCalculators.push_back(new Jclass());
256 }else if (Estimators[i] == "sorclass") {
257 treeCalculators.push_back(new SorClass());
258 }else if (Estimators[i] == "jest") {
259 treeCalculators.push_back(new Jest());
260 }else if (Estimators[i] == "sorest") {
261 treeCalculators.push_back(new SorEst());
262 }else if (Estimators[i] == "thetayc") {
263 treeCalculators.push_back(new ThetaYC());
264 }else if (Estimators[i] == "thetan") {
265 treeCalculators.push_back(new ThetaN());
266 }else if (Estimators[i] == "kstest") {
267 treeCalculators.push_back(new KSTest());
268 }else if (Estimators[i] == "sharednseqs") {
269 treeCalculators.push_back(new SharedNSeqs());
270 }else if (Estimators[i] == "ochiai") {
271 treeCalculators.push_back(new Ochiai());
272 }else if (Estimators[i] == "anderberg") {
273 treeCalculators.push_back(new Anderberg());
274 }else if (Estimators[i] == "kulczynski") {
275 treeCalculators.push_back(new Kulczynski());
276 }else if (Estimators[i] == "kulczynskicody") {
277 treeCalculators.push_back(new KulczynskiCody());
278 }else if (Estimators[i] == "lennon") {
279 treeCalculators.push_back(new Lennon());
280 }else if (Estimators[i] == "morisitahorn") {
281 treeCalculators.push_back(new MorHorn());
282 }else if (Estimators[i] == "braycurtis") {
283 treeCalculators.push_back(new BrayCurtis());
284 }else if (Estimators[i] == "whittaker") {
285 treeCalculators.push_back(new Whittaker());
286 }else if (Estimators[i] == "odum") {
287 treeCalculators.push_back(new Odum());
288 }else if (Estimators[i] == "canberra") {
289 treeCalculators.push_back(new Canberra());
290 }else if (Estimators[i] == "structeuclidean") {
291 treeCalculators.push_back(new StructEuclidean());
292 }else if (Estimators[i] == "structchord") {
293 treeCalculators.push_back(new StructChord());
294 }else if (Estimators[i] == "hellinger") {
295 treeCalculators.push_back(new Hellinger());
296 }else if (Estimators[i] == "manhattan") {
297 treeCalculators.push_back(new Manhattan());
298 }else if (Estimators[i] == "structpearson") {
299 treeCalculators.push_back(new StructPearson());
300 }else if (Estimators[i] == "soergel") {
301 treeCalculators.push_back(new Soergel());
302 }else if (Estimators[i] == "spearman") {
303 treeCalculators.push_back(new Spearman());
304 }else if (Estimators[i] == "structkulczynski") {
305 treeCalculators.push_back(new StructKulczynski());
306 }else if (Estimators[i] == "speciesprofile") {
307 treeCalculators.push_back(new SpeciesProfile());
308 }else if (Estimators[i] == "hamming") {
309 treeCalculators.push_back(new Hamming());
310 }else if (Estimators[i] == "structchi2") {
311 treeCalculators.push_back(new StructChi2());
312 }else if (Estimators[i] == "gower") {
313 treeCalculators.push_back(new Gower());
314 }else if (Estimators[i] == "memchi2") {
315 treeCalculators.push_back(new MemChi2());
316 }else if (Estimators[i] == "memchord") {
317 treeCalculators.push_back(new MemChord());
318 }else if (Estimators[i] == "memeuclidean") {
319 treeCalculators.push_back(new MemEuclidean());
320 }else if (Estimators[i] == "mempearson") {
321 treeCalculators.push_back(new MemPearson());
330 catch(exception& e) {
331 m->errorOut(e, "TreeGroupCommand", "TreeGroupCommand");
336 //**********************************************************************************************************************
338 void TreeGroupCommand::help(){
340 m->mothurOut("The tree.shared command creates a .tre to represent the similiarity between groups or sequences.\n");
341 m->mothurOut("The tree.shared command can only be executed after a successful read.otu command or by providing a distance file.\n");
342 m->mothurOut("The tree.shared command parameters are groups, calc, phylip, column, name, cutoff, precision and label.\n");
343 m->mothurOut("The groups parameter allows you to specify which of the groups in your groupfile you would like included used.\n");
344 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");
345 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");
346 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");
347 m->mothurOut("The tree.shared command should be in the following format: tree.shared(groups=yourGroups, calc=yourCalcs, label=yourLabels).\n");
348 m->mothurOut("Example tree.shared(groups=A-B-C, calc=jabund-sorabund).\n");
349 m->mothurOut("The default value for groups is all the groups in your groupfile.\n");
350 m->mothurOut("The default value for calc is jclass-thetayc.\n");
351 m->mothurOut("The tree.shared command outputs a .tre file for each calculator you specify at each distance you choose.\n");
352 validCalculator->printCalc("treegroup", cout);
353 m->mothurOut("Or the tree.shared command can be in the following format: tree.shared(phylip=yourPhylipFile).\n");
354 m->mothurOut("Example tree.shared(phylip=abrecovery.dist).\n");
355 m->mothurOut("Note: No spaces between parameter labels (i.e. groups), '=' and parameters (i.e.yourGroups).\n\n");
357 catch(exception& e) {
358 m->errorOut(e, "TreeGroupCommand", "help");
364 //**********************************************************************************************************************
366 TreeGroupCommand::~TreeGroupCommand(){
367 globaldata->Groups.clear();
368 if (abort == false) {
370 if (format == "sharedfile") { delete read; delete input; globaldata->ginput = NULL; }
371 else { delete readMatrix; delete matrix; delete list; }
372 delete tmap; globaldata->gTreemap = NULL;
373 delete validCalculator;
378 //**********************************************************************************************************************
380 int TreeGroupCommand::execute(){
383 if (abort == true) { if (calledHelp) { return 0; } return 2; }
385 if (format == "sharedfile") {
386 //if the users entered no valid calculators don't execute command
387 if (treeCalculators.size() == 0) { m->mothurOut("You have given no valid calculators."); m->mothurOutEndLine(); return 0; }
389 if (globaldata->gGroupmap != NULL) { delete globaldata->gGroupmap; globaldata->gGroupmap = NULL; }
391 read = new ReadOTUFile(globaldata->inputFileName);
392 read->read(&*globaldata);
394 input = globaldata->ginput;
395 lookup = input->getSharedRAbundVectors();
396 lastLabel = lookup[0]->getLabel();
398 if (lookup.size() < 2) { m->mothurOut("You have not provided enough valid groups. I cannot run the command."); m->mothurOutEndLine(); return 0; }
400 //used in tree constructor
401 globaldata->runParse = false;
403 //create treemap class from groupmap for tree class to use
404 tmap = new TreeMap();
405 tmap->makeSim(globaldata->gGroupmap);
406 globaldata->gTreemap = tmap;
408 //clear globaldatas old tree names if any
409 globaldata->Treenames.clear();
411 //fills globaldatas tree names
412 globaldata->Treenames = globaldata->Groups;
414 if (m->control_pressed) { return 0; }
419 if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } return 0; }
422 filename = globaldata->inputFileName;
424 if (format == "column") { readMatrix = new ReadColumnMatrix(filename); }
425 else if (format == "phylip") { readMatrix = new ReadPhylipMatrix(filename); }
427 readMatrix->setCutoff(cutoff);
430 nameMap = new NameAssignment(namefile);
437 readMatrix->read(nameMap);
438 list = readMatrix->getListVector();
439 matrix = readMatrix->getMatrix();
442 tmap = new TreeMap();
444 if (m->control_pressed) { return 0; }
447 globaldata->gTreemap = tmap;
449 globaldata->Groups = tmap->namesOfGroups;
451 //clear globaldatas old tree names if any
452 globaldata->Treenames.clear();
454 //fills globaldatas tree names
455 globaldata->Treenames = globaldata->Groups;
457 //used in tree constructor
458 globaldata->runParse = false;
460 if (m->control_pressed) { return 0; }
464 if (m->control_pressed) { return 0; }
466 //create a new filename
467 outputFile = outputDir + m->getRootName(m->getSimpleName(globaldata->inputFileName)) + "tre";
468 outputNames.push_back(outputFile); outputTypes["tree"].push_back(outputFile);
472 if (m->control_pressed) { return 0; }
474 m->mothurOut("Tree complete. "); m->mothurOutEndLine();
478 //reset groups parameter
479 globaldata->Groups.clear();
481 m->mothurOutEndLine();
482 m->mothurOut("Output File Names: "); m->mothurOutEndLine();
483 for (int i = 0; i < outputNames.size(); i++) { m->mothurOut(outputNames[i]); m->mothurOutEndLine(); }
484 m->mothurOutEndLine();
488 catch(exception& e) {
489 m->errorOut(e, "TreeGroupCommand", "execute");
493 //**********************************************************************************************************************
495 int TreeGroupCommand::createTree(){
500 //do merges and create tree structure by setting parents and children
501 //there are numGroups - 1 merges to do
502 for (int i = 0; i < (numGroups - 1); i++) {
503 float largest = -1000.0;
505 if (m->control_pressed) { delete t; return 1; }
508 //find largest value in sims matrix by searching lower triangle
509 for (int j = 1; j < simMatrix.size(); j++) {
510 for (int k = 0; k < j; k++) {
511 if (simMatrix[j][k] > largest) { largest = simMatrix[j][k]; row = j; column = k; }
515 //set non-leaf node info and update leaves to know their parents
517 t->tree[numGroups + i].setChildren(index[row], index[column]);
520 t->tree[index[row]].setParent(numGroups + i);
521 t->tree[index[column]].setParent(numGroups + i);
523 //blength = distance / 2;
524 float blength = ((1.0 - largest) / 2);
527 t->tree[index[row]].setBranchLength(blength - t->tree[index[row]].getLengthToLeaves());
528 t->tree[index[column]].setBranchLength(blength - t->tree[index[column]].getLengthToLeaves());
530 //set your length to leaves to your childs length plus branchlength
531 t->tree[numGroups + i].setLengthToLeaves(t->tree[index[row]].getLengthToLeaves() + t->tree[index[row]].getBranchLength());
535 index[row] = numGroups+i;
536 index[column] = numGroups+i;
538 //remove highest value that caused the merge.
539 simMatrix[row][column] = -1000.0;
540 simMatrix[column][row] = -1000.0;
542 //merge values in simsMatrix
543 for (int n = 0; n < simMatrix.size(); n++) {
544 //row becomes merge of 2 groups
545 simMatrix[row][n] = (simMatrix[row][n] + simMatrix[column][n]) / 2;
546 simMatrix[n][row] = simMatrix[row][n];
548 simMatrix[column][n] = -1000.0;
549 simMatrix[n][column] = -1000.0;
553 //adjust tree to make sure root to tip length is .5
554 int root = t->findRoot();
555 t->tree[root].setBranchLength((0.5 - t->tree[root].getLengthToLeaves()));
560 if (m->control_pressed) { delete t; return 1; }
563 t->createNewickFile(outputFile);
568 if (m->control_pressed) { remove(outputFile.c_str()); outputNames.pop_back(); return 1; }
573 catch(exception& e) {
574 m->errorOut(e, "TreeGroupCommand", "createTree");
578 /***********************************************************/
579 void TreeGroupCommand::printSims(ostream& out) {
582 //output column headers
584 //for (int i = 0; i < lookup.size(); i++) { out << lookup[i]->getGroup() << '\t'; }
588 for (int m = 0; m < simMatrix.size(); m++) {
589 //out << lookup[m]->getGroup() << '\t';
590 for (int n = 0; n < simMatrix.size(); n++) {
591 out << simMatrix[m][n] << '\t';
597 catch(exception& e) {
598 m->errorOut(e, "TreeGroupCommand", "printSims");
602 /***********************************************************/
603 int TreeGroupCommand::makeSimsDist() {
605 numGroups = list->size();
609 for (int g = 0; g < numGroups; g++) { index[g] = g; }
611 //initialize simMatrix
613 simMatrix.resize(numGroups);
614 for (int k = 0; k < simMatrix.size(); k++) {
615 for (int j = 0; j < simMatrix.size(); j++) {
616 simMatrix[k].push_back(0.0);
620 //go through sparse matrix and fill sims
621 //go through each cell in the sparsematrix
622 for(MatData currentCell = matrix->begin(); currentCell != matrix->end(); currentCell++){
623 //similairity = -(distance-1)
624 simMatrix[currentCell->row][currentCell->column] = -(currentCell->dist -1.0);
625 simMatrix[currentCell->column][currentCell->row] = -(currentCell->dist -1.0);
627 if (m->control_pressed) { return 1; }
633 catch(exception& e) {
634 m->errorOut(e, "TreeGroupCommand", "makeSimsDist");
639 /***********************************************************/
640 int TreeGroupCommand::makeSimsShared() {
642 set<string> processedLabels;
643 set<string> userLabels = labels;
645 //as long as you are not at the end of the file or done wih the lines you want
646 while((lookup[0] != NULL) && ((allLines == 1) || (userLabels.size() != 0))) {
647 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; }
649 if(allLines == 1 || labels.count(lookup[0]->getLabel()) == 1){
650 m->mothurOut(lookup[0]->getLabel()); m->mothurOutEndLine();
653 processedLabels.insert(lookup[0]->getLabel());
654 userLabels.erase(lookup[0]->getLabel());
657 if ((m->anyLabelsToProcess(lookup[0]->getLabel(), userLabels, "") == true) && (processedLabels.count(lastLabel) != 1)) {
658 string saveLabel = lookup[0]->getLabel();
660 for (int i = 0; i < lookup.size(); i++) { delete lookup[i]; }
661 lookup = input->getSharedRAbundVectors(lastLabel);
663 m->mothurOut(lookup[0]->getLabel()); m->mothurOutEndLine();
666 processedLabels.insert(lookup[0]->getLabel());
667 userLabels.erase(lookup[0]->getLabel());
669 //restore real lastlabel to save below
670 lookup[0]->setLabel(saveLabel);
673 lastLabel = lookup[0]->getLabel();
675 //get next line to process
676 for (int i = 0; i < lookup.size(); i++) { delete lookup[i]; }
677 lookup = input->getSharedRAbundVectors();
680 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; }
682 //output error messages about any remaining user labels
683 set<string>::iterator it;
684 bool needToRun = false;
685 for (it = userLabels.begin(); it != userLabels.end(); it++) {
686 m->mothurOut("Your file does not include the label " + *it);
687 if (processedLabels.count(lastLabel) != 1) {
688 m->mothurOut(". I will use " + lastLabel + "."); m->mothurOutEndLine();
691 m->mothurOut(". Please refer to " + lastLabel + "."); m->mothurOutEndLine();
695 //run last label if you need to
696 if (needToRun == true) {
697 for (int i = 0; i < lookup.size(); i++) { if (lookup[i] != NULL) { delete lookup[i]; } }
698 lookup = input->getSharedRAbundVectors(lastLabel);
700 m->mothurOut(lookup[0]->getLabel()); m->mothurOutEndLine();
702 for (int i = 0; i < lookup.size(); i++) { delete lookup[i]; }
705 for(int i = 0 ; i < treeCalculators.size(); i++) { delete treeCalculators[i]; }
709 catch(exception& e) {
710 m->errorOut(e, "TreeGroupCommand", "makeSimsShared");
715 /***********************************************************/
716 int TreeGroupCommand::process(vector<SharedRAbundVector*> thisLookup) {
719 vector<SharedRAbundVector*> subset;
720 numGroups = thisLookup.size();
722 //for each calculator
723 for(int i = 0 ; i < treeCalculators.size(); i++) {
724 //initialize simMatrix
726 simMatrix.resize(numGroups);
727 for (int k = 0; k < simMatrix.size(); k++) {
728 for (int j = 0; j < simMatrix.size(); j++) {
729 simMatrix[k].push_back(0.0);
735 for (int g = 0; g < numGroups; g++) { index[g] = g; }
737 //create a new filename
738 outputFile = outputDir + m->getRootName(m->getSimpleName(globaldata->inputFileName)) + treeCalculators[i]->getName() + "." + thisLookup[0]->getLabel() + ".tre";
739 outputNames.push_back(outputFile); outputTypes["tree"].push_back(outputFile);
741 for (int k = 0; k < thisLookup.size(); k++) {
742 for (int l = k; l < thisLookup.size(); l++) {
743 if (k != l) { //we dont need to similiarity of a groups to itself
744 //get estimated similarity between 2 groups
746 subset.clear(); //clear out old pair of sharedrabunds
747 //add new pair of sharedrabunds
748 subset.push_back(thisLookup[k]); subset.push_back(thisLookup[l]);
750 //if this calc needs all groups to calculate the pair load all groups
751 if (treeCalculators[i]->getNeedsAll()) {
752 //load subset with rest of lookup for those calcs that need everyone to calc for a pair
753 for (int w = 0; w < thisLookup.size(); w++) {
754 if ((w != k) && (w != l)) { subset.push_back(thisLookup[w]); }
758 data = treeCalculators[i]->getValues(subset); //saves the calculator outputs
759 //cout << thisLookup[k]->getGroup() << '\t' << thisLookup[l]->getGroup() << '\t' << (1.0 - data[0]) << endl;
760 if (m->control_pressed) { return 1; }
762 //save values in similarity matrix
763 simMatrix[k][l] = data[0];
764 simMatrix[l][k] = data[0];
769 //createdistance file from simMatrix
770 /*string o = outputDir + m->getRootName(m->getSimpleName(globaldata->inputFileName)) + treeCalculators[i]->getName() + "." + thisLookup[0]->getLabel() + ".dist";
772 m->openOutputFile(o, outDist);
773 outDist << simMatrix.size() << endl;
774 for (int k = 0; k < simMatrix.size(); k++) {
775 outDist << thisLookup[k]->getGroup() << '\t';
776 for (int l = 0; l < k; l++) {
777 outDist << (1.0-simMatrix[k][l]) << '\t';
784 if (m->control_pressed) { return 1; }
785 //creates tree from similarity matrix and write out file
788 if (m->control_pressed) { return 1; }
794 catch(exception& e) {
795 m->errorOut(e, "TreeGroupCommand", "process");
799 /***********************************************************/