X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=parsimonycommand.cpp;h=56305e2260d4af19050dfc0832add36c1208d13e;hb=a6c698b20eda3671d22466ab6b98b36331a30804;hp=989ea120447540884daac7f4f5e5496946b86d52;hpb=16bea3130e36addc54e2116dfbcd02d706ebee45;p=mothur.git diff --git a/parsimonycommand.cpp b/parsimonycommand.cpp index 989ea12..56305e2 100644 --- a/parsimonycommand.cpp +++ b/parsimonycommand.cpp @@ -21,22 +21,26 @@ ParsimonyCommand::ParsimonyCommand() { if (randomtree == "") { T = globaldata->gTree; tmap = globaldata->gTreemap; - parsFile = globaldata->getTreeFile() + ".parsimony"; - openOutputFile(parsFile, out); + output = new ColumnFile(globaldata->getTreeFile() + ".parsimony"); sumFile = globaldata->getTreeFile() + ".psummary"; openOutputFile(sumFile, outSum); - distFile = globaldata->getTreeFile() + ".pdistrib"; - openOutputFile(distFile, outDist); - }else { //user wants random distribution savetmap = globaldata->gTreemap; getUserInput(); - parsFile = randomtree + ".rd_parsimony"; - openOutputFile(parsFile, out); + output = new ColumnFile(randomtree); } + //set users groups to analyze + util = new SharedUtil(); + util->setGroups(globaldata->Groups, tmap->namesOfGroups, allGroups, numGroups, "unweighted"); //sets the groups the user wants to analyze + util->getCombos(groupComb, globaldata->Groups, numComp); + globaldata->setGroups(""); + + if (numGroups == 1) { numComp++; groupComb.push_back(allGroups); } + convert(globaldata->getIters(), iters); //how many random trees to generate pars = new Parsimony(tmap); + counter = 0; } catch(exception& e) { @@ -51,119 +55,150 @@ ParsimonyCommand::ParsimonyCommand() { /***********************************************************/ int ParsimonyCommand::execute() { try { + Progress* reading; + reading = new Progress("Comparing to random:", iters); //get pscore for users tree - userData.resize(1,0); //data[0] = pscore. - randomData.resize(1,0); //data[0] = pscore. - - //format output - outDist.setf(ios::fixed, ios::floatfield); outDist.setf(ios::showpoint); - outDist << "RandomTree#" << '\t' << "ParsScore" << endl; - + userData.resize(numComp,0); //data = AB, AC, BC, ABC. + randomData.resize(numComp,0); //data = AB, AC, BC, ABC. + rscoreFreq.resize(numComp); + uscoreFreq.resize(numComp); + rCumul.resize(numComp); + uCumul.resize(numComp); + userTreeScores.resize(numComp); + UScoreSig.resize(numComp); + if (randomtree == "") { //get pscores for users trees for (int i = 0; i < T.size(); i++) { - cout << "Processing tree " << i+1 << endl; - userData = pars->getValues(T[i]); //userData[0] = pscore - //update uscoreFreq - it = uscoreFreq.find(userData[0]); - if (it == uscoreFreq.end()) {//new score - uscoreFreq[userData[0]] = 1; - }else{ uscoreFreq[userData[0]]++; } - - //add users score to valid scores - validScores[userData[0]] = userData[0]; - - //save score for summary file - userTreeScores.push_back(userData[0]); - + userData = pars->getValues(T[i]); //data = AB, AC, BC, ABC. + + //output scores for each combination + for(int k = 0; k < numComp; k++) { + + //update uscoreFreq + it = uscoreFreq[k].find(userData[k]); + if (it == uscoreFreq[k].end()) {//new score + uscoreFreq[k][userData[k]] = 1; + }else{ uscoreFreq[k][userData[k]]++; } + + //add users score to valid scores + validScores[userData[k]] = userData[k]; + + //save score for summary file + userTreeScores[k].push_back(userData[k]); + } } //get pscores for random trees for (int j = 0; j < iters; j++) { //create new tree with same num nodes and leaves as users randT = new Tree(); + //create random relationships between nodes randT->assembleRandomTree(); + //get pscore of random tree randomData = pars->getValues(randT); + + for(int r = 0; r < numComp; r++) { + //add trees pscore to map of scores + it2 = rscoreFreq[r].find(randomData[r]); + if (it2 != rscoreFreq[r].end()) {//already have that score + rscoreFreq[r][randomData[r]]++; + }else{//first time we have seen this score + rscoreFreq[r][randomData[r]] = 1; + } - //add trees pscore to map of scores - it2 = rscoreFreq.find(randomData[0]); - if (it2 != rscoreFreq.end()) {//already have that score - rscoreFreq[randomData[0]]++; - }else{//first time we have seen this score - rscoreFreq[randomData[0]] = 1; + //add randoms score to validscores + validScores[randomData[r]] = randomData[r]; } - - //add randoms score to validscores - validScores[randomData[0]] = randomData[0]; - //output info to pdistrib file - outDist << j+1 << '\t'<< '\t' << randomData[0] << endl; - + //update progress bar + reading->update(j); + delete randT; } + }else { //get pscores for random trees for (int j = 0; j < iters; j++) { //create new tree with same num nodes and leaves as users randT = new Tree(); //create random relationships between nodes + randT->assembleRandomTree(); + //get pscore of random tree randomData = pars->getValues(randT); - //add trees pscore to map of scores - it2 = rscoreFreq.find(randomData[0]); - if (it2 != rscoreFreq.end()) {//already have that score - rscoreFreq[randomData[0]]++; - }else{//first time we have seen this score - rscoreFreq[randomData[0]] = 1; - } + for(int r = 0; r < numComp; r++) { + //add trees pscore to map of scores + it2 = rscoreFreq[r].find(randomData[r]); + if (it2 != rscoreFreq[r].end()) {//already have that score + rscoreFreq[r][randomData[r]]++; + }else{//first time we have seen this score + rscoreFreq[r][randomData[r]] = 1; + } - //add randoms score to validscores - validScores[randomData[0]] = randomData[0]; - + //add randoms score to validscores + validScores[randomData[r]] = randomData[r]; + } + + //update progress bar + reading->update(j); + delete randT; } } - - float rcumul = 0.0000; - float ucumul = 0.0000; - - //this loop fills the cumulative maps and put 0.0000 in the score freq map to make it easier to print. - for (it = validScores.begin(); it != validScores.end(); it++) { - if (randomtree == "") { - it2 = uscoreFreq.find(it->first); - //user data has that score - if (it2 != uscoreFreq.end()) { uscoreFreq[it->first] /= T.size(); ucumul+= it2->second; } - else { uscoreFreq[it->first] = 0.0000; } //no user trees with that score - //make uCumul map - uCumul[it->first] = ucumul; + + for(int a = 0; a < numComp; a++) { + float rcumul = 0.0000; + float ucumul = 0.0000; + //this loop fills the cumulative maps and put 0.0000 in the score freq map to make it easier to print. + for (it = validScores.begin(); it != validScores.end(); it++) { + if (randomtree == "") { + it2 = uscoreFreq[a].find(it->first); + //user data has that score + if (it2 != uscoreFreq[a].end()) { uscoreFreq[a][it->first] /= T.size(); ucumul+= it2->second; } + else { uscoreFreq[a][it->first] = 0.0000; } //no user trees with that score + //make uCumul map + uCumul[a][it->first] = ucumul; + } + + //make rscoreFreq map and rCumul + it2 = rscoreFreq[a].find(it->first); + //get percentage of random trees with that info + if (it2 != rscoreFreq[a].end()) { rscoreFreq[a][it->first] /= iters; rcumul+= it2->second; } + else { rscoreFreq[a][it->first] = 0.0000; } //no random trees with that score + rCumul[a][it->first] = rcumul; } - //make rscoreFreq map and rCumul - it2 = rscoreFreq.find(it->first); - //get percentage of random trees with that info - if (it2 != rscoreFreq.end()) { rscoreFreq[it->first] /= iters; rcumul+= it2->second; } - else { rscoreFreq[it->first] = 0.0000; } //no random trees with that score - rCumul[it->first] = rcumul; + //find the signifigance of each user trees score when compared to the random trees and save for printing the summary file + for (int h = 0; h < userTreeScores[a].size(); h++) { + UScoreSig[a].push_back(rCumul[a][userTreeScores[a][h]]); + } } - //find the signifigance of each user trees score when compared to the random trees and save for printing the summary file - for (int h = 0; h < userTreeScores.size(); h++) { - UScoreSig.push_back(rCumul[userTreeScores[h]]); - } + //finish progress bar + reading->finish(); + delete reading; + printParsimonyFile(); - printUSummaryFile(); + if (randomtree == "") { printUSummaryFile(); } //reset globaldata's treemap if you just did random distrib - if (randomtree != "") { globaldata->gTreemap = savetmap; } + if (randomtree != "") { + //memory leak prevention + //if (globaldata->gTreemap != NULL) { delete globaldata->gTreemap; } + globaldata->gTreemap = savetmap; + } //reset randomTree parameter to "" globaldata->setRandomTree(""); + //reset groups parameter + globaldata->Groups.clear(); return 0; @@ -181,27 +216,29 @@ int ParsimonyCommand::execute() { /***********************************************************/ void ParsimonyCommand::printParsimonyFile() { try { - //column headers + vector data; + vector tags; + if (randomtree == "") { - out << "Score" << '\t' << "UserFreq" << '\t' << "UserCumul" << '\t' << "RandFreq" << '\t' << "RandCumul" << endl; + tags.push_back("Score"); tags.push_back("UserFreq"); tags.push_back("UserCumul"); tags.push_back("RandFreq"); tags.push_back("RandCumul"); }else { - out << "Score" << '\t' << "RandFreq" << '\t' << "RandCumul" << endl; + tags.push_back("Score"); tags.push_back("RandFreq"); tags.push_back("RandCumul"); + } + + for(int a = 0; a < numComp; a++) { + output->initFile(groupComb[a], tags); + //print each line + for (it = validScores.begin(); it != validScores.end(); it++) { + if (randomtree == "") { + data.push_back(it->first); data.push_back(uscoreFreq[a][it->first]); data.push_back(uCumul[a][it->first]); data.push_back(rscoreFreq[a][it->first]); data.push_back(rCumul[a][it->first]); + }else{ + data.push_back(it->first); data.push_back(rscoreFreq[a][it->first]); data.push_back(rCumul[a][it->first]); + } + output->output(data); + data.clear(); + } + output->resetFile(); } - - //format output - out.setf(ios::fixed, ios::floatfield); out.setf(ios::showpoint); - - //print each line - for (it = validScores.begin(); it != validScores.end(); it++) { - if (randomtree == "") { - out << setprecision(6) << it->first << '\t' << '\t' << uscoreFreq[it->first] << '\t' << uCumul[it->first] << '\t' << rscoreFreq[it->first] << '\t' << rCumul[it->first] << endl; - }else{ - out << setprecision(6) << it->first << '\t' << '\t' << rscoreFreq[it->first] << '\t' << rCumul[it->first] << endl; - } - } - - out.close(); - } catch(exception& e) { cout << "Standard Error: " << e.what() << " has occurred in the ParsimonyCommand class Function printParsimonyFile. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; @@ -216,14 +253,24 @@ void ParsimonyCommand::printParsimonyFile() { void ParsimonyCommand::printUSummaryFile() { try { //column headers - outSum << "Tree#" << '\t' << "ParsScore" << '\t' << '\t' << "ParsSig" << endl; + outSum << "Tree#" << '\t' << "Groups" << '\t' << "ParsScore" << '\t' << "ParsSig" << endl; + cout << "Tree#" << '\t' << "Groups" << '\t' << "ParsScore" << '\t' << "ParsSig" << endl; //format output outSum.setf(ios::fixed, ios::floatfield); outSum.setf(ios::showpoint); + //print each line for (int i = 0; i< T.size(); i++) { - outSum << setprecision(6) << i+1 << '\t' << '\t' << userTreeScores[i] << '\t' << UScoreSig[i] << endl; + for(int a = 0; a < numComp; a++) { + if (UScoreSig[a][i] > (1/(float)iters)) { + outSum << setprecision(6) << i+1 << '\t' << groupComb[a] << '\t' << userTreeScores[a][i] << setprecision(globaldata->getIters().length()) << '\t' << UScoreSig[a][i] << endl; + cout << setprecision(6) << i+1 << '\t' << groupComb[a] << '\t' << userTreeScores[a][i] << setprecision(globaldata->getIters().length()) << '\t' << UScoreSig[a][i] << endl; + }else { + outSum << setprecision(6) << i+1 << '\t' << groupComb[a] << '\t' << userTreeScores[a][i] << setprecision(globaldata->getIters().length()) << '\t' << "<" << (1/float(iters)) << endl; + cout << setprecision(6) << i+1 << '\t' << groupComb[a] << '\t' << userTreeScores[a][i] << setprecision(globaldata->getIters().length()) << '\t' << "<" << (1/float(iters)) << endl; + } + } } outSum.close(); @@ -258,6 +305,7 @@ void ParsimonyCommand::getUserInput() { //set tmaps seqsPerGroup tmap->seqsPerGroup[toString(i)] = num; + tmap->namesOfGroups.push_back(toString(i)); //set tmaps namesOfSeqs for (int j = 0; j < num; j++) { @@ -272,6 +320,8 @@ void ParsimonyCommand::getUserInput() { getline(cin, s); //save tmap for later + //memory leak prevention + //if (globaldata->gTreemap != NULL) { delete globaldata->gTreemap; } globaldata->gTreemap = tmap; } @@ -284,5 +334,7 @@ void ParsimonyCommand::getUserInput() { exit(1); } } + /***********************************************************/ +