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 "sharedjabund.h"
12 #include "sharedsorabund.h"
13 #include "sharedjclass.h"
14 #include "sharedsorclass.h"
15 #include "sharedjest.h"
16 #include "sharedsorest.h"
17 #include "sharedthetayc.h"
18 #include "sharedthetan.h"
19 #include "sharedmorisitahorn.h"
20 #include "sharedbraycurtis.h"
23 //**********************************************************************************************************************
25 TreeGroupCommand::TreeGroupCommand(){
27 globaldata = GlobalData::getInstance();
28 format = globaldata->getFormat();
29 validCalculator = new ValidCalculators();
30 util = new SharedUtil();
33 for (i=0; i<globaldata->Estimators.size(); i++) {
34 if (validCalculator->isValidCalculator("treegroup", globaldata->Estimators[i]) == true) {
35 if (globaldata->Estimators[i] == "jabund") {
36 treeCalculators.push_back(new JAbund());
37 }else if (globaldata->Estimators[i] == "sorabund") {
38 treeCalculators.push_back(new SorAbund());
39 }else if (globaldata->Estimators[i] == "jclass") {
40 treeCalculators.push_back(new Jclass());
41 }else if (globaldata->Estimators[i] == "sorclass") {
42 treeCalculators.push_back(new SorClass());
43 }else if (globaldata->Estimators[i] == "jest") {
44 treeCalculators.push_back(new Jest());
45 }else if (globaldata->Estimators[i] == "sorest") {
46 treeCalculators.push_back(new SorEst());
47 }else if (globaldata->Estimators[i] == "thetayc") {
48 treeCalculators.push_back(new ThetaYC());
49 }else if (globaldata->Estimators[i] == "thetan") {
50 treeCalculators.push_back(new ThetaN());
51 }else if (globaldata->Estimators[i] == "morisitahorn") {
52 treeCalculators.push_back(new MorHorn());
53 }else if (globaldata->Estimators[i] == "braycurtis") {
54 treeCalculators.push_back(new BrayCurtis());
59 //reset calc for next command
60 globaldata->setCalc("");
64 cout << "Standard Error: " << e.what() << " has occurred in the TreeGroupCommand class Function TreeGroupCommand. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
68 cout << "An unknown error has occurred in the TreeGroupCommand class function TreeGroupCommand. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
72 //**********************************************************************************************************************
74 TreeGroupCommand::~TreeGroupCommand(){
80 //**********************************************************************************************************************
82 int TreeGroupCommand::execute(){
86 vector<SharedRAbundVector*> subset;
88 //if the users entered no valid calculators don't execute command
89 if (treeCalculators.size() == 0) { return 0; }
91 if (format == "sharedfile") {
93 read = new ReadOTUFile(globaldata->inputFileName);
94 read->read(&*globaldata);
96 input = globaldata->ginput;
97 lookup = input->getSharedRAbundVectors();
99 //you are using a list and a groupfile
100 read = new ReadOTUFile(globaldata->inputFileName);
101 read->read(&*globaldata);
103 input = globaldata->ginput;
104 SharedList = globaldata->gSharedList;
105 lookup = SharedList->getSharedRAbundVector();
108 if (lookup.size() < 2) { cout << "You have not provided enough valid groups. I cannot run the command." << endl; }
110 numGroups = globaldata->Groups.size();
112 for (int i = 0; i < numGroups; i++) { groupNames += globaldata->Groups[i]; }
114 //clear globaldatas old tree names if any
115 globaldata->Treenames.clear();
117 //fills globaldatas tree names
118 globaldata->Treenames = globaldata->Groups;
120 //create treemap class from groupmap for tree class to use
121 tmap = new TreeMap();
122 tmap->makeSim(globaldata->gGroupmap);
123 globaldata->gTreemap = tmap;
125 while(lookup[0] != NULL){
127 if(globaldata->allLines == 1 || globaldata->lines.count(count) == 1 || globaldata->labels.count(lookup[0]->getLabel()) == 1){
129 cout << lookup[0]->getLabel() << '\t' << count << endl;
131 //for each calculator
132 for(int i = 0 ; i < treeCalculators.size(); i++) {
134 //initialize simMatrix
136 simMatrix.resize(numGroups);
137 for (int m = 0; m < simMatrix.size(); m++) {
138 for (int j = 0; j < simMatrix.size(); j++) {
139 simMatrix[m].push_back(0.0);
145 for (int g = 0; g < numGroups; g++) { index[g] = g; }
147 //create a new filename
148 outputFile = getRootName(globaldata->inputFileName) + treeCalculators[i]->getName() + "." + lookup[0]->getLabel() + ".tre";
150 for (int k = 0; k < lookup.size(); k++) {
151 for (int l = k; l < lookup.size(); l++) {
152 if (k != l) { //we dont need to similiarity of a groups to itself
153 //get estimated similarity between 2 groups
155 subset.clear(); //clear out old pair of sharedrabunds
156 //add new pair of sharedrabunds
157 subset.push_back(lookup[k]); subset.push_back(lookup[l]);
159 data = treeCalculators[i]->getValues(subset); //saves the calculator outputs
160 //save values in similarity matrix
161 simMatrix[k][l] = data[0];
162 simMatrix[l][k] = data[0];
167 //creates tree from similarity matrix and write out file
172 //get next line to process
173 lookup = input->getSharedRAbundVectors();
177 //reset groups parameter
178 globaldata->Groups.clear(); globaldata->setGroups("");
182 catch(exception& e) {
183 cout << "Standard Error: " << e.what() << " has occurred in the TreeGroupCommand class Function execute. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
187 cout << "An unknown error has occurred in the TreeGroupCommand class function execute. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
191 //**********************************************************************************************************************
193 void TreeGroupCommand::createTree(){
198 //do merges and create tree structure by setting parents and children
199 //there are numGroups - 1 merges to do
200 for (int i = 0; i < (numGroups - 1); i++) {
204 //find largest value in sims matrix by searching lower triangle
205 for (int j = 1; j < simMatrix.size(); j++) {
206 for (int k = 0; k < j; k++) {
207 if (simMatrix[j][k] > largest) { largest = simMatrix[j][k]; row = j; column = k; }
211 //set non-leaf node info and update leaves to know their parents
213 t->tree[numGroups + i].setChildren(index[row], index[column]);
216 t->tree[index[row]].setParent(numGroups + i);
217 t->tree[index[column]].setParent(numGroups + i);
219 //blength = distance / 2;
220 float blength = ((1.0 - largest) / 2);
223 t->tree[index[row]].setBranchLength(blength - t->tree[index[row]].getLengthToLeaves());
224 t->tree[index[column]].setBranchLength(blength - t->tree[index[column]].getLengthToLeaves());
226 //set your length to leaves to your childs length plus branchlength
227 t->tree[numGroups + i].setLengthToLeaves(t->tree[index[row]].getLengthToLeaves() + t->tree[index[row]].getBranchLength());
231 index[row] = numGroups+i;
232 index[column] = numGroups+i;
234 //zero out highest value that caused the merge.
235 simMatrix[row][column] = 0.0;
236 simMatrix[column][row] = 0.0;
238 //merge values in simsMatrix
239 for (int n = 0; n < simMatrix.size(); n++) {
240 //row becomes merge of 2 groups
241 simMatrix[row][n] = (simMatrix[row][n] + simMatrix[column][n]) / 2;
242 simMatrix[n][row] = simMatrix[row][n];
244 simMatrix[column][n] = 0.0;
245 simMatrix[n][column] = 0.0;
253 t->createNewickFile(outputFile);
259 catch(exception& e) {
260 cout << "Standard Error: " << e.what() << " has occurred in the TreeGroupCommand class Function createTree. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
264 cout << "An unknown error has occurred in the TreeGroupCommand class function createTree. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
268 /***********************************************************/
269 void TreeGroupCommand::printSims() {
271 cout << "simsMatrix" << endl;
272 for (int m = 0; m < simMatrix.size(); m++) {
273 for (int n = 0; n < simMatrix.size(); n++) {
274 cout << simMatrix[m][n] << '\t';
280 catch(exception& e) {
281 cout << "Standard Error: " << e.what() << " has occurred in the TreeGroupCommand class Function printSims. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
285 cout << "An unknown error has occurred in the TreeGroupCommand class function printSims. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
289 /***********************************************************/