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::setParameters(){
56 CommandParameter pshared("shared", "InputTypes", "", "", "PhylipColumnShared", "PhylipColumnShared", "none",false,false); parameters.push_back(pshared);
57 CommandParameter pphylip("phylip", "InputTypes", "", "", "PhylipColumnShared", "PhylipColumnShared", "none",false,false); parameters.push_back(pphylip);
58 CommandParameter pname("name", "InputTypes", "", "", "none", "none", "ColumnName",false,false); parameters.push_back(pname);
59 CommandParameter pcolumn("column", "InputTypes", "", "", "PhylipColumnShared", "PhylipColumnShared", "ColumnName",false,false); parameters.push_back(pcolumn);
60 CommandParameter pcutoff("cutoff", "Number", "", "10", "", "", "",false,false); parameters.push_back(pcutoff);
61 CommandParameter pprecision("precision", "Number", "", "100", "", "", "",false,false); parameters.push_back(pprecision);
62 CommandParameter plabel("label", "String", "", "", "", "", "",false,false); parameters.push_back(plabel);
63 CommandParameter pgroups("groups", "String", "", "", "", "", "",false,false); parameters.push_back(pgroups);
64 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);
65 CommandParameter poutput("output", "Multiple", "lt-square", "lt", "", "", "",false,false); parameters.push_back(poutput);
66 CommandParameter pinputdir("inputdir", "String", "", "", "", "", "",false,false); parameters.push_back(pinputdir);
67 CommandParameter poutputdir("outputdir", "String", "", "", "", "", "",false,false); parameters.push_back(poutputdir);
69 vector<string> myArray;
70 for (int i = 0; i < parameters.size(); i++) { myArray.push_back(parameters[i].name); }
74 m->errorOut(e, "TreeGroupCommand", "setParameters");
78 //**********************************************************************************************************************
79 string TreeGroupCommand::getHelpString(){
81 string helpString = "";
82 ValidCalculators validCalculator;
83 helpString += "The tree.shared command creates a .tre to represent the similiarity between groups or sequences.\n";
84 helpString += "The tree.shared command parameters are shared, groups, calc, phylip, column, name, cutoff, precision and label.\n";
85 helpString += "The groups parameter allows you to specify which of the groups in your groupfile you would like included used.\n";
86 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";
87 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";
88 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";
89 helpString += "The tree.shared command should be in the following format: tree.shared(groups=yourGroups, calc=yourCalcs, label=yourLabels).\n";
90 helpString += "Example tree.shared(groups=A-B-C, calc=jabund-sorabund).\n";
91 helpString += "The default value for groups is all the groups in your groupfile.\n";
92 helpString += "The default value for calc is jclass-thetayc.\n";
93 helpString += "The tree.shared command outputs a .tre file for each calculator you specify at each distance you choose.\n";
94 helpString += validCalculator.printCalc("treegroup");
95 helpString += "Or the tree.shared command can be in the following format: tree.shared(phylip=yourPhylipFile).\n";
96 helpString += "Example tree.shared(phylip=abrecovery.dist).\n";
97 helpString += "Note: No spaces between parameter labels (i.e. groups), '=' and parameters (i.e.yourGroups).\n";
100 catch(exception& e) {
101 m->errorOut(e, "TreeGroupCommand", "getHelpString");
105 //**********************************************************************************************************************
106 TreeGroupCommand::TreeGroupCommand(){
108 abort = true; calledHelp = true;
110 //initialize outputTypes
111 vector<string> tempOutNames;
112 outputTypes["tree"] = tempOutNames;
114 catch(exception& e) {
115 m->errorOut(e, "TreeGroupCommand", "TreeGroupCommand");
119 //**********************************************************************************************************************
121 TreeGroupCommand::TreeGroupCommand(string option) {
123 abort = false; calledHelp = false;
126 //allow user to run help
127 if(option == "help") { help(); abort = true; calledHelp = true; }
130 vector<string> myArray = setParameters();
132 OptionParser parser(option);
133 map<string, string> parameters = parser. getParameters();
135 ValidParameters validParameter;
136 map<string, string>::iterator it;
138 //check to make sure all parameters are valid for command
139 for (it = parameters.begin(); it != parameters.end(); it++) {
140 if (validParameter.isValidParameter(it->first, myArray, it->second) != true) { abort = true; }
143 //initialize outputTypes
144 vector<string> tempOutNames;
145 outputTypes["tree"] = tempOutNames;
147 //if the user changes the input directory command factory will send this info to us in the output parameter
148 string inputDir = validParameter.validFile(parameters, "inputdir", false);
149 if (inputDir == "not found"){ inputDir = ""; }
152 it = parameters.find("phylip");
153 //user has given a template file
154 if(it != parameters.end()){
155 path = m->hasPath(it->second);
156 //if the user has not given a path then, add inputdir. else leave path alone.
157 if (path == "") { parameters["phylip"] = inputDir + it->second; }
160 it = parameters.find("column");
161 //user has given a template file
162 if(it != parameters.end()){
163 path = m->hasPath(it->second);
164 //if the user has not given a path then, add inputdir. else leave path alone.
165 if (path == "") { parameters["column"] = inputDir + it->second; }
168 it = parameters.find("name");
169 //user has given a template file
170 if(it != parameters.end()){
171 path = m->hasPath(it->second);
172 //if the user has not given a path then, add inputdir. else leave path alone.
173 if (path == "") { parameters["name"] = inputDir + it->second; }
177 //check for required parameters
178 phylipfile = validParameter.validFile(parameters, "phylip", true);
179 if (phylipfile == "not open") { phylipfile = ""; abort = true; }
180 else if (phylipfile == "not found") { phylipfile = ""; }
181 else { inputfile = phylipfile; format = "phylip"; }
183 columnfile = validParameter.validFile(parameters, "column", true);
184 if (columnfile == "not open") { columnfile = ""; abort = true; }
185 else if (columnfile == "not found") { columnfile = ""; }
186 else { inputfile = columnfile; format = "column"; }
188 sharedfile = validParameter.validFile(parameters, "shared", true);
189 if (sharedfile == "not open") { sharedfile = ""; abort = true; }
190 else if (sharedfile == "not found") { sharedfile = ""; }
191 else { inputfile = sharedfile; format = "sharedfile"; }
193 namefile = validParameter.validFile(parameters, "name", true);
194 if (namefile == "not open") { abort = true; }
195 else if (namefile == "not found") { namefile = ""; }
197 if ((phylipfile == "") && (columnfile == "") && (sharedfile == "")) {
198 //is there are current file available for either of these?
199 //give priority to shared, then column, then phylip
200 sharedfile = m->getSharedFile();
201 if (sharedfile != "") { inputfile = sharedfile; format = "sharedfile"; m->mothurOut("Using " + sharedfile + " as input file for the shared parameter."); m->mothurOutEndLine(); }
203 columnfile = m->getColumnFile();
204 if (columnfile != "") { inputfile = columnfile; format = "column"; m->mothurOut("Using " + columnfile + " as input file for the column parameter."); m->mothurOutEndLine(); }
206 phylipfile = m->getPhylipFile();
207 if (phylipfile != "") { inputfile = phylipfile; format = "phylip"; m->mothurOut("Using " + phylipfile + " as input file for the phylip parameter."); m->mothurOutEndLine(); }
209 m->mothurOut("No valid current files. You must provide a shared, phylip or column file."); m->mothurOutEndLine();
215 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; }
217 if (columnfile != "") {
218 if (namefile == "") {
219 namefile = m->getNameFile();
220 if (namefile != "") { m->mothurOut("Using " + namefile + " as input file for the name parameter."); m->mothurOutEndLine(); }
222 m->mothurOut("You need to provide a namefile if you are going to use the column format."); m->mothurOutEndLine();
228 //check for optional parameter and set defaults
229 // ...at some point should added some additional type checking...
230 label = validParameter.validFile(parameters, "label", false);
231 if (label == "not found") { label = ""; }
233 if(label != "all") { m->splitAtDash(label, labels); allLines = 0; }
234 else { allLines = 1; }
237 groups = validParameter.validFile(parameters, "groups", false);
238 if (groups == "not found") { groups = ""; }
240 m->splitAtDash(groups, Groups);
244 calc = validParameter.validFile(parameters, "calc", false);
245 if (calc == "not found") { calc = "jclass-thetayc"; }
247 if (calc == "default") { calc = "jclass-thetayc"; }
249 m->splitAtDash(calc, Estimators);
252 temp = validParameter.validFile(parameters, "precision", false); if (temp == "not found") { temp = "100"; }
253 convert(temp, precision);
255 temp = validParameter.validFile(parameters, "cutoff", false); if (temp == "not found") { temp = "10"; }
256 convert(temp, cutoff);
257 cutoff += (5 / (precision * 10.0));
259 //if the user changes the output directory command factory will send this info to us in the output parameter
260 outputDir = validParameter.validFile(parameters, "outputdir", false); if (outputDir == "not found"){
262 outputDir += m->hasPath(inputfile); //if user entered a file with a path then preserve it
267 catch(exception& e) {
268 m->errorOut(e, "TreeGroupCommand", "TreeGroupCommand");
272 //**********************************************************************************************************************
274 TreeGroupCommand::~TreeGroupCommand(){
275 if (abort == false) {
276 if (format == "sharedfile") { delete input; }
277 else { delete readMatrix; delete matrix; delete list; }
283 //**********************************************************************************************************************
285 int TreeGroupCommand::execute(){
288 if (abort == true) { if (calledHelp) { return 0; } return 2; }
290 if (format == "sharedfile") {
292 ValidCalculators validCalculator;
294 for (int i=0; i<Estimators.size(); i++) {
295 if (validCalculator.isValidCalculator("treegroup", Estimators[i]) == true) {
296 if (Estimators[i] == "sharedsobs") {
297 treeCalculators.push_back(new SharedSobsCS());
298 }else if (Estimators[i] == "sharedchao") {
299 treeCalculators.push_back(new SharedChao1());
300 }else if (Estimators[i] == "sharedace") {
301 treeCalculators.push_back(new SharedAce());
302 }else if (Estimators[i] == "jabund") {
303 treeCalculators.push_back(new JAbund());
304 }else if (Estimators[i] == "sorabund") {
305 treeCalculators.push_back(new SorAbund());
306 }else if (Estimators[i] == "jclass") {
307 treeCalculators.push_back(new Jclass());
308 }else if (Estimators[i] == "sorclass") {
309 treeCalculators.push_back(new SorClass());
310 }else if (Estimators[i] == "jest") {
311 treeCalculators.push_back(new Jest());
312 }else if (Estimators[i] == "sorest") {
313 treeCalculators.push_back(new SorEst());
314 }else if (Estimators[i] == "thetayc") {
315 treeCalculators.push_back(new ThetaYC());
316 }else if (Estimators[i] == "thetan") {
317 treeCalculators.push_back(new ThetaN());
318 }else if (Estimators[i] == "kstest") {
319 treeCalculators.push_back(new KSTest());
320 }else if (Estimators[i] == "sharednseqs") {
321 treeCalculators.push_back(new SharedNSeqs());
322 }else if (Estimators[i] == "ochiai") {
323 treeCalculators.push_back(new Ochiai());
324 }else if (Estimators[i] == "anderberg") {
325 treeCalculators.push_back(new Anderberg());
326 }else if (Estimators[i] == "kulczynski") {
327 treeCalculators.push_back(new Kulczynski());
328 }else if (Estimators[i] == "kulczynskicody") {
329 treeCalculators.push_back(new KulczynskiCody());
330 }else if (Estimators[i] == "lennon") {
331 treeCalculators.push_back(new Lennon());
332 }else if (Estimators[i] == "morisitahorn") {
333 treeCalculators.push_back(new MorHorn());
334 }else if (Estimators[i] == "braycurtis") {
335 treeCalculators.push_back(new BrayCurtis());
336 }else if (Estimators[i] == "whittaker") {
337 treeCalculators.push_back(new Whittaker());
338 }else if (Estimators[i] == "odum") {
339 treeCalculators.push_back(new Odum());
340 }else if (Estimators[i] == "canberra") {
341 treeCalculators.push_back(new Canberra());
342 }else if (Estimators[i] == "structeuclidean") {
343 treeCalculators.push_back(new StructEuclidean());
344 }else if (Estimators[i] == "structchord") {
345 treeCalculators.push_back(new StructChord());
346 }else if (Estimators[i] == "hellinger") {
347 treeCalculators.push_back(new Hellinger());
348 }else if (Estimators[i] == "manhattan") {
349 treeCalculators.push_back(new Manhattan());
350 }else if (Estimators[i] == "structpearson") {
351 treeCalculators.push_back(new StructPearson());
352 }else if (Estimators[i] == "soergel") {
353 treeCalculators.push_back(new Soergel());
354 }else if (Estimators[i] == "spearman") {
355 treeCalculators.push_back(new Spearman());
356 }else if (Estimators[i] == "structkulczynski") {
357 treeCalculators.push_back(new StructKulczynski());
358 }else if (Estimators[i] == "speciesprofile") {
359 treeCalculators.push_back(new SpeciesProfile());
360 }else if (Estimators[i] == "hamming") {
361 treeCalculators.push_back(new Hamming());
362 }else if (Estimators[i] == "structchi2") {
363 treeCalculators.push_back(new StructChi2());
364 }else if (Estimators[i] == "gower") {
365 treeCalculators.push_back(new Gower());
366 }else if (Estimators[i] == "memchi2") {
367 treeCalculators.push_back(new MemChi2());
368 }else if (Estimators[i] == "memchord") {
369 treeCalculators.push_back(new MemChord());
370 }else if (Estimators[i] == "memeuclidean") {
371 treeCalculators.push_back(new MemEuclidean());
372 }else if (Estimators[i] == "mempearson") {
373 treeCalculators.push_back(new MemPearson());
378 //if the users entered no valid calculators don't execute command
379 if (treeCalculators.size() == 0) { m->mothurOut("You have given no valid calculators."); m->mothurOutEndLine(); return 0; }
381 input = new InputData(sharedfile, "sharedfile");
382 lookup = input->getSharedRAbundVectors();
383 lastLabel = lookup[0]->getLabel();
385 if (lookup.size() < 2) { m->mothurOut("You have not provided enough valid groups. I cannot run the command."); m->mothurOutEndLine(); return 0; }
387 //used in tree constructor
390 //create treemap class from groupmap for tree class to use
391 tmap = new TreeMap();
392 tmap->makeSim(m->namesOfGroups);
394 //clear globaldatas old tree names if any
395 m->Treenames.clear();
397 //fills globaldatas tree names
398 m->Treenames = m->Groups;
400 if (m->control_pressed) { return 0; }
405 if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } return 0; }
408 filename = inputfile;
410 if (format == "column") { readMatrix = new ReadColumnMatrix(filename); }
411 else if (format == "phylip") { readMatrix = new ReadPhylipMatrix(filename); }
413 readMatrix->setCutoff(cutoff);
416 nameMap = new NameAssignment(namefile);
423 readMatrix->read(nameMap);
424 list = readMatrix->getListVector();
425 matrix = readMatrix->getMatrix();
428 tmap = new TreeMap();
430 if (m->control_pressed) { return 0; }
434 m->Groups = tmap->namesOfGroups;
436 //clear globaldatas old tree names if any
437 m->Treenames.clear();
439 //fills globaldatas tree names
440 m->Treenames = m->Groups;
442 //used in tree constructor
445 if (m->control_pressed) { return 0; }
449 if (m->control_pressed) { return 0; }
451 //create a new filename
452 outputFile = outputDir + m->getRootName(m->getSimpleName(inputfile)) + "tre";
453 outputNames.push_back(outputFile); outputTypes["tree"].push_back(outputFile);
457 if (m->control_pressed) { return 0; }
459 m->mothurOut("Tree complete. "); m->mothurOutEndLine();
463 //reset groups parameter
466 //set tree file as new current treefile
468 itTypes = outputTypes.find("tree");
469 if (itTypes != outputTypes.end()) {
470 if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setTreeFile(current); }
473 m->mothurOutEndLine();
474 m->mothurOut("Output File Names: "); m->mothurOutEndLine();
475 for (int i = 0; i < outputNames.size(); i++) { m->mothurOut(outputNames[i]); m->mothurOutEndLine(); }
476 m->mothurOutEndLine();
480 catch(exception& e) {
481 m->errorOut(e, "TreeGroupCommand", "execute");
485 //**********************************************************************************************************************
487 int TreeGroupCommand::createTree(){
492 //do merges and create tree structure by setting parents and children
493 //there are numGroups - 1 merges to do
494 for (int i = 0; i < (numGroups - 1); i++) {
495 float largest = -1000.0;
497 if (m->control_pressed) { delete t; return 1; }
500 //find largest value in sims matrix by searching lower triangle
501 for (int j = 1; j < simMatrix.size(); j++) {
502 for (int k = 0; k < j; k++) {
503 if (simMatrix[j][k] > largest) { largest = simMatrix[j][k]; row = j; column = k; }
507 //set non-leaf node info and update leaves to know their parents
509 t->tree[numGroups + i].setChildren(index[row], index[column]);
512 t->tree[index[row]].setParent(numGroups + i);
513 t->tree[index[column]].setParent(numGroups + i);
515 //blength = distance / 2;
516 float blength = ((1.0 - largest) / 2);
519 t->tree[index[row]].setBranchLength(blength - t->tree[index[row]].getLengthToLeaves());
520 t->tree[index[column]].setBranchLength(blength - t->tree[index[column]].getLengthToLeaves());
522 //set your length to leaves to your childs length plus branchlength
523 t->tree[numGroups + i].setLengthToLeaves(t->tree[index[row]].getLengthToLeaves() + t->tree[index[row]].getBranchLength());
527 index[row] = numGroups+i;
528 index[column] = numGroups+i;
530 //remove highest value that caused the merge.
531 simMatrix[row][column] = -1000.0;
532 simMatrix[column][row] = -1000.0;
534 //merge values in simsMatrix
535 for (int n = 0; n < simMatrix.size(); n++) {
536 //row becomes merge of 2 groups
537 simMatrix[row][n] = (simMatrix[row][n] + simMatrix[column][n]) / 2;
538 simMatrix[n][row] = simMatrix[row][n];
540 simMatrix[column][n] = -1000.0;
541 simMatrix[n][column] = -1000.0;
545 //adjust tree to make sure root to tip length is .5
546 int root = t->findRoot();
547 t->tree[root].setBranchLength((0.5 - t->tree[root].getLengthToLeaves()));
552 if (m->control_pressed) { delete t; return 1; }
555 t->createNewickFile(outputFile);
560 if (m->control_pressed) { remove(outputFile.c_str()); outputNames.pop_back(); return 1; }
565 catch(exception& e) {
566 m->errorOut(e, "TreeGroupCommand", "createTree");
570 /***********************************************************/
571 void TreeGroupCommand::printSims(ostream& out) {
574 //output column headers
576 //for (int i = 0; i < lookup.size(); i++) { out << lookup[i]->getGroup() << '\t'; }
580 for (int m = 0; m < simMatrix.size(); m++) {
581 //out << lookup[m]->getGroup() << '\t';
582 for (int n = 0; n < simMatrix.size(); n++) {
583 out << simMatrix[m][n] << '\t';
589 catch(exception& e) {
590 m->errorOut(e, "TreeGroupCommand", "printSims");
594 /***********************************************************/
595 int TreeGroupCommand::makeSimsDist() {
597 numGroups = list->size();
601 for (int g = 0; g < numGroups; g++) { index[g] = g; }
603 //initialize simMatrix
605 simMatrix.resize(numGroups);
606 for (int k = 0; k < simMatrix.size(); k++) {
607 for (int j = 0; j < simMatrix.size(); j++) {
608 simMatrix[k].push_back(0.0);
612 //go through sparse matrix and fill sims
613 //go through each cell in the sparsematrix
614 for(MatData currentCell = matrix->begin(); currentCell != matrix->end(); currentCell++){
615 //similairity = -(distance-1)
616 simMatrix[currentCell->row][currentCell->column] = -(currentCell->dist -1.0);
617 simMatrix[currentCell->column][currentCell->row] = -(currentCell->dist -1.0);
619 if (m->control_pressed) { return 1; }
625 catch(exception& e) {
626 m->errorOut(e, "TreeGroupCommand", "makeSimsDist");
631 /***********************************************************/
632 int TreeGroupCommand::makeSimsShared() {
634 set<string> processedLabels;
635 set<string> userLabels = labels;
637 //as long as you are not at the end of the file or done wih the lines you want
638 while((lookup[0] != NULL) && ((allLines == 1) || (userLabels.size() != 0))) {
639 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; }
641 if(allLines == 1 || labels.count(lookup[0]->getLabel()) == 1){
642 m->mothurOut(lookup[0]->getLabel()); m->mothurOutEndLine();
645 processedLabels.insert(lookup[0]->getLabel());
646 userLabels.erase(lookup[0]->getLabel());
649 if ((m->anyLabelsToProcess(lookup[0]->getLabel(), userLabels, "") == true) && (processedLabels.count(lastLabel) != 1)) {
650 string saveLabel = lookup[0]->getLabel();
652 for (int i = 0; i < lookup.size(); i++) { delete lookup[i]; }
653 lookup = input->getSharedRAbundVectors(lastLabel);
655 m->mothurOut(lookup[0]->getLabel()); m->mothurOutEndLine();
658 processedLabels.insert(lookup[0]->getLabel());
659 userLabels.erase(lookup[0]->getLabel());
661 //restore real lastlabel to save below
662 lookup[0]->setLabel(saveLabel);
665 lastLabel = lookup[0]->getLabel();
667 //get next line to process
668 for (int i = 0; i < lookup.size(); i++) { delete lookup[i]; }
669 lookup = input->getSharedRAbundVectors();
672 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; }
674 //output error messages about any remaining user labels
675 set<string>::iterator it;
676 bool needToRun = false;
677 for (it = userLabels.begin(); it != userLabels.end(); it++) {
678 m->mothurOut("Your file does not include the label " + *it);
679 if (processedLabels.count(lastLabel) != 1) {
680 m->mothurOut(". I will use " + lastLabel + "."); m->mothurOutEndLine();
683 m->mothurOut(". Please refer to " + lastLabel + "."); m->mothurOutEndLine();
687 //run last label if you need to
688 if (needToRun == true) {
689 for (int i = 0; i < lookup.size(); i++) { if (lookup[i] != NULL) { delete lookup[i]; } }
690 lookup = input->getSharedRAbundVectors(lastLabel);
692 m->mothurOut(lookup[0]->getLabel()); m->mothurOutEndLine();
694 for (int i = 0; i < lookup.size(); i++) { delete lookup[i]; }
697 for(int i = 0 ; i < treeCalculators.size(); i++) { delete treeCalculators[i]; }
701 catch(exception& e) {
702 m->errorOut(e, "TreeGroupCommand", "makeSimsShared");
707 /***********************************************************/
708 int TreeGroupCommand::process(vector<SharedRAbundVector*> thisLookup) {
711 vector<SharedRAbundVector*> subset;
712 numGroups = thisLookup.size();
714 //for each calculator
715 for(int i = 0 ; i < treeCalculators.size(); i++) {
716 //initialize simMatrix
718 simMatrix.resize(numGroups);
719 for (int k = 0; k < simMatrix.size(); k++) {
720 for (int j = 0; j < simMatrix.size(); j++) {
721 simMatrix[k].push_back(0.0);
727 for (int g = 0; g < numGroups; g++) { index[g] = g; }
729 //create a new filename
730 outputFile = outputDir + m->getRootName(m->getSimpleName(inputfile)) + treeCalculators[i]->getName() + "." + thisLookup[0]->getLabel() + ".tre";
731 outputNames.push_back(outputFile); outputTypes["tree"].push_back(outputFile);
733 for (int k = 0; k < thisLookup.size(); k++) {
734 for (int l = k; l < thisLookup.size(); l++) {
735 if (k != l) { //we dont need to similiarity of a groups to itself
736 //get estimated similarity between 2 groups
738 subset.clear(); //clear out old pair of sharedrabunds
739 //add new pair of sharedrabunds
740 subset.push_back(thisLookup[k]); subset.push_back(thisLookup[l]);
742 //if this calc needs all groups to calculate the pair load all groups
743 if (treeCalculators[i]->getNeedsAll()) {
744 //load subset with rest of lookup for those calcs that need everyone to calc for a pair
745 for (int w = 0; w < thisLookup.size(); w++) {
746 if ((w != k) && (w != l)) { subset.push_back(thisLookup[w]); }
750 data = treeCalculators[i]->getValues(subset); //saves the calculator outputs
751 //cout << thisLookup[k]->getGroup() << '\t' << thisLookup[l]->getGroup() << '\t' << (1.0 - data[0]) << endl;
752 if (m->control_pressed) { return 1; }
754 //save values in similarity matrix
755 simMatrix[k][l] = data[0];
756 simMatrix[l][k] = data[0];
761 //createdistance file from simMatrix
762 /*string o = outputDir + m->getRootName(m->getSimpleName(globaldata->inputFileName)) + treeCalculators[i]->getName() + "." + thisLookup[0]->getLabel() + ".dist";
764 m->openOutputFile(o, outDist);
765 outDist << simMatrix.size() << endl;
766 for (int k = 0; k < simMatrix.size(); k++) {
767 outDist << thisLookup[k]->getGroup() << '\t';
768 for (int l = 0; l < k; l++) {
769 outDist << (1.0-simMatrix[k][l]) << '\t';
776 if (m->control_pressed) { return 1; }
777 //creates tree from similarity matrix and write out file
780 if (m->control_pressed) { return 1; }
786 catch(exception& e) {
787 m->errorOut(e, "TreeGroupCommand", "process");
791 /***********************************************************/