X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=unifracweightedcommand.cpp;h=2652f8f6de6cc1198e81112c2d7e3d88d417fad3;hb=9651e8e7172d86707b34af15e95ec60ad4c3c3f9;hp=47b23d667e7b159531b208a49f80666bc7ba48c0;hpb=fb5a52d8d7896ba00c9ccc6add22ed89c5fb2ad7;p=mothur.git diff --git a/unifracweightedcommand.cpp b/unifracweightedcommand.cpp index 47b23d6..2652f8f 100644 --- a/unifracweightedcommand.cpp +++ b/unifracweightedcommand.cpp @@ -16,53 +16,10 @@ UnifracWeightedCommand::UnifracWeightedCommand() { T = globaldata->gTree; tmap = globaldata->gTreemap; - weightedFile = globaldata->getTreeFile() + ".weighted"; - openOutputFile(weightedFile, out); sumFile = globaldata->getTreeFile() + ".wsummary"; openOutputFile(sumFile, outSum); - distFile = globaldata->getTreeFile() + ".wdistrib"; - openOutputFile(distFile, outDist); - - //if the user has not entered specific groups to analyze then do them all - if (globaldata->Groups.size() == 0) { - numGroups = tmap->getNumGroups(); - }else { - //check that groups are valid - for (int i = 0; i < globaldata->Groups.size(); i++) { - if (tmap->isValidGroup(globaldata->Groups[i]) != true) { - cout << globaldata->Groups[i] << " is not a valid group, and will be disregarded." << endl; - // erase the invalid group from globaldata->Groups - globaldata->Groups.erase (globaldata->Groups.begin()+i); - } - } - - //if the user only entered invalid groups - if (globaldata->Groups.size() == 0) { - numGroups = tmap->getNumGroups(); - cout << "When using the groups parameter you must have at least 2 valid groups. I will run the command using all the groups in your groupfile." << endl; - }else if (globaldata->Groups.size() == 1) { - cout << "When using the groups parameter you must have at least 2 valid groups. I will run the command using all the groups in your groupfile." << endl; - numGroups = tmap->getNumGroups(); - globaldata->Groups.clear(); - }else { numGroups = globaldata->Groups.size(); } - } - - //calculate number of comparisons i.e. with groups A,B,C = AB, AC, BC = 3; - numComp = 0; - int n = 1; - for (int i=1; iGroups.size() != 0) { - groupComb.push_back(globaldata->Groups[i-1]+globaldata->Groups[l]); - }else { - groupComb.push_back(tmap->namesOfGroups[i-1]+tmap->namesOfGroups[l]); - } - } - n++; - } - + + setGroups(); //sets the groups the user wants to analyze convert(globaldata->getIters(), iters); //how many random trees to generate weighted = new Weighted(tmap); @@ -83,123 +40,77 @@ int UnifracWeightedCommand::execute() { //get weighted for users tree userData.resize(numComp,0); //data[0] = weightedscore AB, data[1] = weightedscore AC... randomData.resize(numComp,0); //data[0] = weightedscore AB, data[1] = weightedscore AC... - uscoreFreq.resize(numComp); - validScores.resize(numComp); - totalrscoreFreq.resize(numComp); - uCumul.resize(numComp); - - //format output - outDist.setf(ios::fixed, ios::floatfield); outDist.setf(ios::showpoint); - outDist << "Tree#" << '\t' << "Iter" << '\t' << "Groups"<< '\t' << "WScore" << endl; - - + //create new tree with same num nodes and leaves as users randT = new Tree(); - //get pscores for users trees + //get weighted scores for users trees for (int i = 0; i < T.size(); i++) { - rscoreFreq.resize(numComp); //data[0] = weightedscore AB, data[1] = weightedscore AC... - rCumul.resize(numComp); //data[0] = weightedscore AB, data[1] = weightedscore AC... - - cout << "Processing tree " << i+1 << endl; + counter = 0; + rScores.resize(numComp); //data[0] = weightedscore AB, data[1] = weightedscore AC... + uScores.resize(numComp); //data[0] = weightedscore AB, data[1] = weightedscore AC... + weightedFile = globaldata->getTreeFile() + toString(i+1) + ".weighted"; + weightedFileout = globaldata->getTreeFile() + "temp." + toString(i+1) + ".weighted"; + userData = weighted->getValues(T[i]); //userData[0] = weightedscore //save users score for (int s=0; sgetCopy(T[i]); - //get scores for random trees for (int j = 0; j < iters; j++) { - //create a random tree with same topology as T[i], but different labels - randT->assembleRandomUnifracTree(); - //get pscore of random tree - randomData = weighted->getValues(randT); - - //save ramdoms score - for (int p=0; pgetCopy(T[i]); + + //create a random tree with same topology as T[i], but different labels + randT->assembleRandomUnifracTree(globaldata->Groups[r], globaldata->Groups[l]); + //get wscore of random tree + randomData = weighted->getValues(randT, globaldata->Groups[r], globaldata->Groups[l]); + + //save scores + rScores[count].push_back(randomData[0]); + count++; } - - //add random score to valid scores - validScores[p][randomData[p]] = randomData[p]; - - //output info to uwdistrib file - outDist << i+1 << '\t' << '\t'<< j+1 << '\t' << '\t' << groupComb[p] << '\t'<< randomData[p] << endl; } } - - saveRandomScores(); //save all random scores for weighted file - + + //removeValidScoresDuplicates(); //find the signifigance of the score for summary file - for (int t = 0; t < numComp; t++) { - float rcumul = 1.0000; - for (it = validScores[t].begin(); it != validScores[t].end(); it++) { - //make rscoreFreq map and rCumul - it2 = rscoreFreq[t].find(it->first); - rCumul[t][it->first] = rcumul; - //get percentage of random trees with that info - if (it2 != rscoreFreq[t].end()) { rscoreFreq[t][it->first] /= iters; rcumul-= it2->second; } - else { rscoreFreq[t][it->first] = 0.0000; } //no random trees with that score - } - } - - //save the signifigance of the users score for printing later for (int f = 0; f < numComp; f++) { - WScoreSig.push_back(rCumul[f][userData[f]]); - } + //sort random scores + sort(rScores[f].begin(), rScores[f].end()); + + //the index of the score higher than yours is returned + //so if you have 1000 random trees the index returned is 100 + //then there are 900 trees with a score greater then you. + //giving you a signifigance of 0.900 + int index = findIndex(userData[f], f); if (index == -1) { cout << "error in UnifracWeightedCommand" << endl; exit(1); } //error code + //the signifigance is the number of trees with the users score or higher + WScoreSig.push_back((iters-index)/(float)iters); + } - //clear random data - rscoreFreq.clear(); - rCumul.clear(); - } - - rCumul.resize(numComp); - for (int b = 0; b < numComp; b++) { - float ucumul = 1.0000; - float rcumul = 1.0000; - //this loop fills the cumulative maps and put 0.0000 in the score freq map to make it easier to print. - for (it = validScores[b].begin(); it != validScores[b].end(); it++) { - it2 = uscoreFreq[b].find(it->first); - //make uCumul map - uCumul[b][it->first] = ucumul; - //user data has that score - if (it2 != uscoreFreq[b].end()) { uscoreFreq[b][it->first] /= T.size(); ucumul-= it2->second; } - else { uscoreFreq[b][it->first] = 0.0000; } //no user trees with that score + //out << "Tree# " << i << endl; + calculateFreqsCumuls(); + printWeightedFile(); - //make rscoreFreq map and rCumul - it2 = totalrscoreFreq[b].find(it->first); - rCumul[b][it->first] = rcumul; - //get percentage of random trees with that info - if (it2 != totalrscoreFreq[b].end()) { totalrscoreFreq[b][it->first] /= (iters * T.size()); rcumul-= it2->second; } - else { totalrscoreFreq[b][it->first] = 0.0000; } //no random trees with that score - } + //clear data + rScores.clear(); + uScores.clear(); + validScores.clear(); } - printWeightedFile(); printWSummaryFile(); - //reset randomTree parameter to 0 - globaldata->setRandomTree("0"); //clear out users groups globaldata->Groups.clear(); @@ -220,22 +131,22 @@ int UnifracWeightedCommand::execute() { /***********************************************************/ void UnifracWeightedCommand::printWeightedFile() { try { - //column headers + vector data; - out << "Group" << '\t' << "Score" << '\t' << "UserFreq" << '\t' << "UserCumul" << '\t' << "RandFreq" << '\t' << "RandCumul" << endl; - - //format output - out.setf(ios::fixed, ios::floatfield); out.setf(ios::showpoint); - - //for each group - for (int e = 0; e < numComp; e++) { - //print each line in that group - for (it = validScores[e].begin(); it != validScores[e].end(); it++) { - out << setprecision(6) << groupComb[e] << '\t' << it->first << '\t' << '\t' << uscoreFreq[e][it->first] << '\t' << uCumul[e][it->first] << '\t' << totalrscoreFreq[e][it->first] << '\t' << rCumul[e][it->first] << endl; + for(int a = 0; a < numComp; a++) { + initFile(groupComb[a]); + //print each line + for (it = validScores.begin(); it != validScores.end(); it++) { + data.push_back(it->first); data.push_back(rScoreFreq[a][it->first]); data.push_back(rCumul[a][it->first]); + output(data); + data.clear(); } + resetFile(); } out.close(); + inFile.close(); + remove(weightedFileout.c_str()); } catch(exception& e) { @@ -253,7 +164,8 @@ void UnifracWeightedCommand::printWeightedFile() { void UnifracWeightedCommand::printWSummaryFile() { try { //column headers - outSum << "Tree#" << '\t' << "Groups" << '\t' << '\t' << "WScore" << '\t' << '\t' << "WSig" << endl; + outSum << "Tree#" << '\t' << "Groups" << '\t' << "WScore" << '\t' << "WSig" << endl; + cout << "Tree#" << '\t' << "Groups" << '\t' << "WScore" << '\t' << "WSig" << endl; //format output outSum.setf(ios::fixed, ios::floatfield); outSum.setf(ios::showpoint); @@ -262,7 +174,13 @@ void UnifracWeightedCommand::printWSummaryFile() { int count = 0; for (int i = 0; i < T.size(); i++) { for (int j = 0; j < numComp; j++) { - outSum << setprecision(6) << i+1 << '\t' << '\t' << groupComb[j] << '\t' << utreeScores[count] << '\t' << WScoreSig[count] << endl; + if (WScoreSig[count] > (1/(float)iters)) { + outSum << setprecision(6) << i+1 << '\t' << '\t' << groupComb[j] << '\t' << '\t' << utreeScores[count] << '\t' << setprecision(globaldata->getIters().length()) << WScoreSig[count] << endl; + cout << setprecision(6) << i+1 << '\t' << '\t' << groupComb[j] << '\t' << '\t' << utreeScores[count] << '\t' << setprecision(globaldata->getIters().length()) << WScoreSig[count] << endl; + }else{ + outSum << setprecision(6) << i+1 << '\t' << groupComb[j] << '\t' << utreeScores[count] << '\t' << setprecision(globaldata->getIters().length()) << "<" << (1/float(iters)) << endl; + cout << setprecision(6) << i+1 << '\t' << groupComb[j] << '\t' << utreeScores[count] << '\t' << setprecision(globaldata->getIters().length()) << "<" << (1/float(iters)) << endl; + } count++; } } @@ -279,30 +197,217 @@ void UnifracWeightedCommand::printWSummaryFile() { } /***********************************************************/ -void UnifracWeightedCommand::saveRandomScores() { +int UnifracWeightedCommand::findIndex(float score, int index) { + try{ + for (int i = 0; i < rScores[index].size(); i++) { + if (rScores[index][i] >= score) { return i; } + } + return rScores[index].size(); + } + catch(exception& e) { + cout << "Standard Error: " << e.what() << " has occurred in the UnifracWeightedCommand class Function findIndex. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + exit(1); + } + catch(...) { + cout << "An unknown error has occurred in the UnifracWeightedCommand class function findIndex. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + exit(1); + } +} + +/***********************************************************/ +void UnifracWeightedCommand::setGroups() { try { - for (int e = 0; e < numComp; e++) { - //update total map with new random scores - for (it = rscoreFreq[e].begin(); it != rscoreFreq[e].end(); it++) { - //does this score already exist in the total map - it2 = totalrscoreFreq[e].find(it->first); - //if yes then add them - if (it2 != totalrscoreFreq[e].end()) { - totalrscoreFreq[e][it->first] += rscoreFreq[e][it->first]; - }else{ //its a new score - totalrscoreFreq[e][it->first] = rscoreFreq[e][it->first]; + //if the user has not entered specific groups to analyze then do them all + if (globaldata->Groups.size() == 0) { + numGroups = tmap->getNumGroups(); + for (int i=0; i < numGroups; i++) { + globaldata->Groups.push_back(tmap->namesOfGroups[i]); + } + }else { + if (globaldata->getGroups() != "all") { + //check that groups are valid + for (int i = 0; i < globaldata->Groups.size(); i++) { + if (tmap->isValidGroup(globaldata->Groups[i]) != true) { + cout << globaldata->Groups[i] << " is not a valid group, and will be disregarded." << endl; + // erase the invalid group from globaldata->Groups + globaldata->Groups.erase (globaldata->Groups.begin()+i); + } + } + + //if the user only entered invalid groups + if (globaldata->Groups.size() == 0) { + numGroups = tmap->getNumGroups(); + for (int i=0; i < numGroups; i++) { + globaldata->Groups.push_back(tmap->namesOfGroups[i]); + } + cout << "When using the groups parameter you must have at least 2 valid groups. I will run the command using all the groups in your groupfile." << endl; + }else if (globaldata->Groups.size() == 1) { + cout << "When using the groups parameter you must have at least 2 valid groups. I will run the command using all the groups in your groupfile." << endl; + numGroups = tmap->getNumGroups(); + globaldata->Groups.clear(); + for (int i=0; i < numGroups; i++) { + globaldata->Groups.push_back(tmap->namesOfGroups[i]); + } + }else { numGroups = globaldata->Groups.size(); } + }else { //users wants all groups + numGroups = tmap->getNumGroups(); + globaldata->Groups.clear(); + globaldata->setGroups(""); + for (int i=0; i < numGroups; i++) { + globaldata->Groups.push_back(tmap->namesOfGroups[i]); } } } + + //calculate number of comparisons i.e. with groups A,B,C = AB, AC, BC = 3; + numComp = 0; + for (int i=0; iGroups[i] + "-" + globaldata->Groups[l]); + } + } } catch(exception& e) { - cout << "Standard Error: " << e.what() << " has occurred in the UnifracWeightedCommand class Function saveRandomScores. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + cout << "Standard Error: " << e.what() << " has occurred in the UnifracWeightedCommand class Function setGroups. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; exit(1); } catch(...) { - cout << "An unknown error has occurred in the UnifracWeightedCommand class function saveRandomScores. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + cout << "An unknown error has occurred in the UnifracWeightedCommand class function setGroups. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; exit(1); } } /***********************************************************/ + +void UnifracWeightedCommand::calculateFreqsCumuls() { + try { + //clear out old tree values + rScoreFreq.clear(); + rScoreFreq.resize(numComp); + rCumul.clear(); + rCumul.resize(numComp); + validScores.clear(); + + //calculate frequency + for (int f = 0; f < numComp; f++) { + for (int i = 0; i < rScores[f].size(); i++) { //looks like 0,0,1,1,1,2,4,7... you want to make a map that say rScoreFreq[0] = 2, rScoreFreq[1] = 3... + validScores[rScores[f][i]] = rScores[f][i]; + it = rScoreFreq[f].find(rScores[f][i]); + if (it != rScoreFreq[f].end()) { + rScoreFreq[f][rScores[f][i]]++; + }else{ + rScoreFreq[f][rScores[f][i]] = 1; + } + } + } + + //calculate rcumul + for(int a = 0; a < numComp; a++) { + float rcumul = 1.0000; + //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++) { + //make rscoreFreq map and rCumul + it2 = rScoreFreq[a].find(it->first); + rCumul[a][it->first] = rcumul; + //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 + } + } + + } + catch(exception& e) { + cout << "Standard Error: " << e.what() << " has occurred in the UnifracWeightedCommand class Function calculateFreqsCums. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + exit(1); + } + catch(...) { + cout << "An unknown error has occurred in the UnifracWeightedCommand class function calculateFreqsCums. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + exit(1); + } + +} + +/*****************************************************************/ + +void UnifracWeightedCommand::initFile(string label){ + try { + if(counter != 0){ + openOutputFile(weightedFileout, out); + openInputFile(weightedFile, inFile); + + string inputBuffer; + getline(inFile, inputBuffer); + + out << inputBuffer << '\t' << label + "RandFreq" << '\t' << label + "RandCumul" << endl; + }else{ + openOutputFile(weightedFileout, out); + out << label + "Score" << '\t' << label + "RandFreq" << '\t' << label + "RandCumul" << endl; + } + + out.setf(ios::fixed, ios::floatfield); + out.setf(ios::showpoint); + } + catch(exception& e) { + cout << "Standard Error: " << e.what() << " has occurred in the UnifracWeightedCommand class Function initFile. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + exit(1); + } + catch(...) { + cout << "An unknown error has occurred in the UnifracWeightedCommand class function initFile. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + exit(1); + } +} + +/***********************************************************************/ + +void UnifracWeightedCommand::output(vector data){ + try { + if(counter != 0){ + string inputBuffer; + getline(inFile, inputBuffer); + + out << inputBuffer << '\t' << setprecision(6) << data[0] << setprecision(globaldata->getIters().length()) << '\t' << data[1] << '\t' << data[2] << endl; + } + else{ + out << setprecision(6) << data[0] << setprecision(globaldata->getIters().length()) << '\t' << data[1] << '\t' << data[2] << endl; + + } + + } + catch(exception& e) { + cout << "Standard Error: " << e.what() << " has occurred in the UnifracWeightedCommand class Function output. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + exit(1); + } + catch(...) { + cout << "An unknown error has occurred in the UnifracWeightedCommand class function output. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + exit(1); + } +}; + +/***********************************************************************/ + +void UnifracWeightedCommand::resetFile(){ + try { + if(counter != 0){ + out.close(); + inFile.close(); + } + else{ + out.close(); + } + counter = 1; + + remove(weightedFile.c_str()); + rename(weightedFileout.c_str(), weightedFile.c_str()); + } + catch(exception& e) { + cout << "Standard Error: " << e.what() << " has occurred in the UnifracWeightedCommand class Function resetFile. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + exit(1); + } + catch(...) { + cout << "An unknown error has occurred in the UnifracWeightedCommand class function resetFile. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + exit(1); + } +} +