5 * Created by Sarah Westcott on 1/26/09.
6 * Copyright 2009 Schloss Lab UMASS Amherst. All rights reserved.
10 #include "parsimonycommand.h"
12 /***********************************************************/
13 ParsimonyCommand::ParsimonyCommand() {
15 globaldata = GlobalData::getInstance();
17 //randomtree will tell us if user had their own treefile or if they just want the random distribution
18 randomtree = globaldata->getRandomTree();
20 //user has entered their own tree
21 if (randomtree == "") {
22 T = globaldata->gTree;
23 tmap = globaldata->gTreemap;
24 parsFile = globaldata->getTreeFile() + ".parsimony";
25 openOutputFile(parsFile, out);
26 sumFile = globaldata->getTreeFile() + ".psummary";
27 openOutputFile(sumFile, outSum);
28 }else { //user wants random distribution
29 savetmap = globaldata->gTreemap;
31 parsFile = randomtree + ".rd_parsimony";
32 openOutputFile(parsFile, out);
35 //set users groups to analyze
37 convert(globaldata->getIters(), iters); //how many random trees to generate
38 pars = new Parsimony(tmap);
42 cout << "Standard Error: " << e.what() << " has occurred in the ParsimonyCommand class Function ParsimonyCommand. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
46 cout << "An unknown error has occurred in the ParsimonyCommand class function ParsimonyCommand. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
50 /***********************************************************/
51 int ParsimonyCommand::execute() {
54 //get pscore for users tree
55 userData.resize(numComp,0); //data = AB, AC, BC, ABC.
56 randomData.resize(numComp,0); //data = AB, AC, BC, ABC.
57 rscoreFreq.resize(numComp);
58 uscoreFreq.resize(numComp);
59 rCumul.resize(numComp);
60 uCumul.resize(numComp);
61 validScores.resize(numComp);
62 userTreeScores.resize(numComp);
63 UScoreSig.resize(numComp);
65 if (randomtree == "") {
66 //get pscores for users trees
67 for (int i = 0; i < T.size(); i++) {
68 cout << "Processing tree " << i+1 << endl;
69 userData = pars->getValues(T[i]); //data = AB, AC, BC, ABC.
71 //output scores for each combination
72 for(int k = 0; k < numComp; k++) {
74 it = uscoreFreq[k].find(userData[k]);
75 if (it == uscoreFreq[k].end()) {//new score
76 uscoreFreq[k][userData[k]] = 1;
77 }else{ uscoreFreq[k][userData[k]]++; }
79 //add users score to valid scores
80 validScores[k][userData[k]] = userData[k];
82 //save score for summary file
83 userTreeScores[k].push_back(userData[k]);
87 //get pscores for random trees
88 for (int j = 0; j < iters; j++) {
89 //create new tree with same num nodes and leaves as users
91 //create random relationships between nodes
92 randT->assembleRandomTree();
93 //get pscore of random tree
94 randomData = pars->getValues(randT);
96 for(int r = 0; r < numComp; r++) {
97 //add trees pscore to map of scores
98 it2 = rscoreFreq[r].find(randomData[r]);
99 if (it2 != rscoreFreq[r].end()) {//already have that score
100 rscoreFreq[r][randomData[r]]++;
101 }else{//first time we have seen this score
102 rscoreFreq[r][randomData[r]] = 1;
105 //add randoms score to validscores
106 validScores[r][randomData[r]] = randomData[r];
112 //get pscores for random trees
113 for (int j = 0; j < iters; j++) {
114 //create new tree with same num nodes and leaves as users
116 //create random relationships between nodes
117 randT->assembleRandomTree();
118 //get pscore of random tree
119 randomData = pars->getValues(randT);
121 for(int r = 0; r < numComp; r++) {
122 //add trees pscore to map of scores
123 it2 = rscoreFreq[r].find(randomData[r]);
124 if (it2 != rscoreFreq[r].end()) {//already have that score
125 rscoreFreq[r][randomData[r]]++;
126 }else{//first time we have seen this score
127 rscoreFreq[r][randomData[r]] = 1;
130 //add randoms score to validscores
131 validScores[r][randomData[r]] = randomData[r];
138 float rcumul = 0.0000;
139 float ucumul = 0.0000;
141 for(int a = 0; a < numComp; a++) {
142 //this loop fills the cumulative maps and put 0.0000 in the score freq map to make it easier to print.
143 for (it = validScores[a].begin(); it != validScores[a].end(); it++) {
144 if (randomtree == "") {
145 it2 = uscoreFreq[a].find(it->first);
146 //user data has that score
147 if (it2 != uscoreFreq[a].end()) { uscoreFreq[a][it->first] /= T.size(); ucumul+= it2->second; }
148 else { uscoreFreq[a][it->first] = 0.0000; } //no user trees with that score
150 uCumul[a][it->first] = ucumul-a;
153 //make rscoreFreq map and rCumul
154 it2 = rscoreFreq[a].find(it->first);
155 //get percentage of random trees with that info
156 if (it2 != rscoreFreq[a].end()) { rscoreFreq[a][it->first] /= iters; rcumul+= it2->second; }
157 else { rscoreFreq[a][it->first] = 0.0000; } //no random trees with that score
158 rCumul[a][it->first] = rcumul-a;
161 //find the signifigance of each user trees score when compared to the random trees and save for printing the summary file
162 for (int h = 0; h < userTreeScores[a].size(); h++) {
163 UScoreSig[a].push_back(rCumul[a][userTreeScores[a][h]]);
167 printParsimonyFile();
168 if (randomtree == "") { printUSummaryFile(); }
170 //reset globaldata's treemap if you just did random distrib
171 if (randomtree != "") { globaldata->gTreemap = savetmap; }
173 //reset randomTree parameter to ""
174 globaldata->setRandomTree("");
175 //reset groups parameter
176 globaldata->Groups.clear(); globaldata->setGroups("");
181 catch(exception& e) {
182 cout << "Standard Error: " << e.what() << " has occurred in the ParsimonyCommand class Function execute. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
186 cout << "An unknown error has occurred in the ParsimonyCommand class function execute. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
191 /***********************************************************/
192 void ParsimonyCommand::printParsimonyFile() {
195 if (randomtree == "") {
196 out << "Comb" << '\t' << "Score" << '\t' << "UserFreq" << '\t' << "UserCumul" << '\t' << "RandFreq" << '\t' << "RandCumul" << endl;
198 out << "Comb" << '\t' << "Score" << '\t' << "RandFreq" << '\t' << "RandCumul" << endl;
202 out.setf(ios::fixed, ios::floatfield); out.setf(ios::showpoint);
204 for(int a = 0; a < numComp; a++) {
206 for (it = validScores[a].begin(); it != validScores[a].end(); it++) {
207 if (randomtree == "") {
208 out << setprecision(6) << groupComb[a] << '\t' << it->first << '\t' << '\t'<< uscoreFreq[a][it->first] << '\t' << uCumul[a][it->first] << '\t' << rscoreFreq[a][it->first] << '\t' << rCumul[a][it->first] << endl;
210 out << setprecision(6) << groupComb[a] << '\t' << it->first << '\t' << '\t' << rscoreFreq[a][it->first] << '\t' << rCumul[a][it->first] << endl;
217 catch(exception& e) {
218 cout << "Standard Error: " << e.what() << " has occurred in the ParsimonyCommand class Function printParsimonyFile. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
222 cout << "An unknown error has occurred in the ParsimonyCommand class function printParsimonyFile. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
226 /***********************************************************/
227 void ParsimonyCommand::printUSummaryFile() {
230 outSum << "Tree#" << '\t' << "Comb" << '\t' << "ParsScore" << '\t' << '\t' << "ParsSig" << endl;
233 outSum.setf(ios::fixed, ios::floatfield); outSum.setf(ios::showpoint);
237 for (int i = 0; i< T.size(); i++) {
238 for(int a = 0; a < numComp; a++) {
239 outSum << setprecision(6) << i+1 << '\t' << groupComb[a] << '\t' << '\t' << userTreeScores[a][i] << '\t' << UScoreSig[a][i] << endl;
240 cout << setprecision(6) << i+1 << '\t' << groupComb[a] << '\t' << '\t' << userTreeScores[a][i] << '\t' << UScoreSig[a][i] << endl;
246 catch(exception& e) {
247 cout << "Standard Error: " << e.what() << " has occurred in the ParsimonyCommand class Function printUSummaryFile. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
251 cout << "An unknown error has occurred in the ParsimonyCommand class function printUSummaryFile. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
256 /***********************************************************/
257 void ParsimonyCommand::getUserInput() {
261 tmap = new TreeMap();
263 cout << "Please enter the number of groups you would like to analyze: ";
268 numEachGroup.resize(numGroups, 0);
270 for (int i = 1; i <= numGroups; i++) {
271 cout << "Please enter the number of sequences in group " << i << ": ";
274 //set tmaps seqsPerGroup
275 tmap->seqsPerGroup[toString(i)] = num;
276 tmap->namesOfGroups.push_back(toString(i));
278 //set tmaps namesOfSeqs
279 for (int j = 0; j < num; j++) {
280 tmap->namesOfSeqs.push_back(toString(count));
281 tmap->treemap[toString(count)].groupname = toString(i);
286 //clears buffer so next command doesn't have error
290 //save tmap for later
291 globaldata->gTreemap = tmap;
294 catch(exception& e) {
295 cout << "Standard Error: " << e.what() << " has occurred in the ParsimonyCommand class Function getUserInput. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
299 cout << "An unknown error has occurred in the ParsimonyCommand class function getUserInput. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
303 /***********************************************************/
305 void ParsimonyCommand::setGroups() {
307 string allGroups = "";
309 //if the user has not entered specific groups to analyze then do them all
310 if (globaldata->Groups.size() != 0) {
311 if (globaldata->Groups[0] != "all") {
312 //check that groups are valid
313 for (int i = 0; i < globaldata->Groups.size(); i++) {
314 if (tmap->isValidGroup(globaldata->Groups[i]) != true) {
315 cout << globaldata->Groups[i] << " is not a valid group, and will be disregarded." << endl;
316 // erase the invalid group from globaldata->Groups
317 globaldata->Groups.erase(globaldata->Groups.begin()+i);
321 //if the user only entered invalid groups
322 if (globaldata->Groups.size() == 0) {
323 cout << "When using the groups parameter you must have at least 1 valid group. I will run the command using all the groups in your groupfile." << endl;
324 for (int i = 0; i < tmap->namesOfGroups.size(); i++) {
325 globaldata->Groups.push_back(tmap->namesOfGroups[i]);
327 allGroups += tmap->namesOfGroups[i];
330 for (int i = 0; i < globaldata->Groups.size(); i++) {
331 allGroups += tmap->namesOfGroups[i];
335 }else{//user has enter "all" and wants the default groups
336 for (int i = 0; i < tmap->namesOfGroups.size(); i++) {
337 globaldata->Groups.push_back(tmap->namesOfGroups[i]);
339 allGroups += tmap->namesOfGroups[i];
341 globaldata->setGroups("");
344 for (int i = 0; i < tmap->namesOfGroups.size(); i++) {
345 allGroups += tmap->namesOfGroups[i];
350 //calculate number of comparsions
352 for (int r=0; r<numGroups; r++) {
353 for (int l = r+1; l < numGroups; l++) {
354 groupComb.push_back(globaldata->Groups[r]+globaldata->Groups[l]);
361 groupComb.push_back(allGroups);
366 catch(exception& e) {
367 cout << "Standard Error: " << e.what() << " has occurred in the ParsimonyCommand class Function setGroups. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
371 cout << "An unknown error has occurred in the ParsimonyCommand class function setGroups. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
376 /*****************************************************************/