X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=parsimonycommand.cpp;h=b87e9c0b1ea0f5e84cbdf87a341c2fc6d393699d;hb=aa9238c0a9e6e7aa0ed8b8b606b08ad4fd7dcfe3;hp=fe5a1dd251899818715cb52c14ac7c73a40630b2;hpb=70ac1dc59244e259b1a972385d490d2ac7f511c0;p=mothur.git diff --git a/parsimonycommand.cpp b/parsimonycommand.cpp index fe5a1dd..b87e9c0 100644 --- a/parsimonycommand.cpp +++ b/parsimonycommand.cpp @@ -10,197 +10,338 @@ #include "parsimonycommand.h" /***********************************************************/ -ParsimonyCommand::ParsimonyCommand() { +ParsimonyCommand::ParsimonyCommand(string option) { try { globaldata = GlobalData::getInstance(); + abort = false; + Groups.clear(); + + //allow user to run help + if(option == "help") { help(); abort = true; } - //randomtree will tell us if user had their own treefile or if they just want the random distribution - randomtree = globaldata->getRandomTree(); + else { + //valid paramters for this command + string Array[] = {"random","groups","iters","outputdir","inputdir"}; + vector myArray (Array, Array+(sizeof(Array)/sizeof(string))); + + OptionParser parser(option); + map parameters = parser.getParameters(); + + ValidParameters validParameter; - //user has entered their own tree - if (randomtree == "") { - T = globaldata->gTree; - tmap = globaldata->gTreemap; - parsFile = globaldata->getTreeFile() + ".parsimony"; - openOutputFile(parsFile, out); - sumFile = globaldata->getTreeFile() + ".psummary"; - openOutputFile(sumFile, outSum); - distFile = globaldata->getTreeFile() + ".pdistrib"; - openOutputFile(distFile, outDist); + //check to make sure all parameters are valid for command + for (map::iterator it = parameters.begin(); it != parameters.end(); it++) { + if (validParameter.isValidParameter(it->first, myArray, it->second) != true) { abort = true; } + } - //if the user has not entered specific groups to analyze then do them all - if (globaldata->Groups.size() != 0) { - //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); - } - } + randomtree = validParameter.validFile(parameters, "random", false); if (randomtree == "not found") { randomtree = ""; } - //if the user only entered invalid groups - if (globaldata->Groups.size() == 0) { - 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; - for (int i = 0; i < tmap->namesOfGroups.size(); i++) { - globaldata->Groups.push_back(tmap->namesOfGroups[i]); - } - } - }else { - for (int i = 0; i < tmap->namesOfGroups.size(); i++) { - globaldata->Groups.push_back(tmap->namesOfGroups[i]); + //are you trying to use parsimony without reading a tree or saying you want random distribution + if (randomtree == "") { + if (globaldata->gTree.size() == 0) { + m->mothurOut("You must read a treefile and a groupfile or set the randomtree parameter to the output filename you wish, before you may execute the parsimony command."); m->mothurOutEndLine(); abort = true; } + } + + //if the user changes the output directory command factory will send this info to us in the output parameter + string outputDir = validParameter.validFile(parameters, "outputdir", false); if (outputDir == "not found"){ outputDir = ""; } + + //check for optional parameter and set defaults + // ...at some point should added some additional type checking... + groups = validParameter.validFile(parameters, "groups", false); + if (groups == "not found") { groups = ""; globaldata->Groups.clear(); } + else { + splitAtDash(groups, Groups); + globaldata->Groups = Groups; + } + + itersString = validParameter.validFile(parameters, "iters", false); if (itersString == "not found") { itersString = "1000"; } + convert(itersString, iters); + + if (abort == false) { + //randomtree will tell us if user had their own treefile or if they just want the random distribution + //user has entered their own tree + if (randomtree == "") { + T = globaldata->gTree; + tmap = globaldata->gTreemap; + + if(outputDir == "") { outputDir += hasPath(globaldata->getTreeFile()); } + output = new ColumnFile(outputDir + getSimpleName(globaldata->getTreeFile()) + ".parsimony", itersString); + outputNames.push_back(outputDir + getSimpleName(globaldata->getTreeFile()) + ".parsimony"); + + sumFile = outputDir + getSimpleName(globaldata->getTreeFile()) + ".psummary"; + openOutputFile(sumFile, outSum); + outputNames.push_back(sumFile); + }else { //user wants random distribution + savetmap = globaldata->gTreemap; + getUserInput(); + + if(outputDir == "") { outputDir += hasPath(randomtree); } + output = new ColumnFile(outputDir+ getSimpleName(randomtree), itersString); + outputNames.push_back(outputDir+ getSimpleName(randomtree)); } + + //set users groups to analyze + util = new SharedUtil(); + util->setGroups(globaldata->Groups, tmap->namesOfGroups, allGroups, numGroups, "parsimony"); //sets the groups the user wants to analyze + util->getCombos(groupComb, globaldata->Groups, numComp); + + if (numGroups == 1) { numComp++; groupComb.push_back(allGroups); } + + pars = new Parsimony(tmap); + counter = 0; + } - - }else { //user wants random distribution - savetmap = globaldata->gTreemap; - getUserInput(); - parsFile = randomtree + ".rd_parsimony"; - openOutputFile(parsFile, out); + } - - convert(globaldata->getIters(), iters); //how many random trees to generate - pars = new Parsimony(tmap); } catch(exception& e) { - cout << "Standard Error: " << e.what() << " has occurred in the ParsimonyCommand class Function ParsimonyCommand. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + m->errorOut(e, "ParsimonyCommand", "ParsimonyCommand"); exit(1); } - catch(...) { - cout << "An unknown error has occurred in the ParsimonyCommand class function ParsimonyCommand. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; +} + +//********************************************************************************************************************** + +void ParsimonyCommand::help(){ + try { + m->mothurOut("The parsimony command can only be executed after a successful read.tree command, unless you use the random parameter.\n"); + m->mothurOut("The parsimony command parameters are random, groups and iters. No parameters are required.\n"); + m->mothurOut("The groups parameter allows you to specify which of the groups in your groupfile you would like analyzed. You must enter at least 1 valid group.\n"); + m->mothurOut("The group names are separated by dashes. The iters parameter allows you to specify how many random trees you would like compared to your tree.\n"); + m->mothurOut("The parsimony command should be in the following format: parsimony(random=yourOutputFilename, groups=yourGroups, iters=yourIters).\n"); + m->mothurOut("Example parsimony(random=out, iters=500).\n"); + m->mothurOut("The default value for random is "" (meaning you want to use the trees in your inputfile, randomtree=out means you just want the random distribution of trees outputted to out.rd_parsimony),\n"); + m->mothurOut("and iters is 1000. The parsimony command output two files: .parsimony and .psummary their descriptions are in the manual.\n"); + m->mothurOut("Note: No spaces between parameter labels (i.e. random), '=' and parameters (i.e.yourOutputFilename).\n\n"); + } + catch(exception& e) { + m->errorOut(e, "ParsimonyCommand", "help"); exit(1); } } + + /***********************************************************/ int ParsimonyCommand::execute() { try { + + if (abort == true) { return 0; } + + 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; + if (m->control_pressed) { + delete reading; delete pars; delete util; delete output; + if (randomtree == "") { outSum.close(); } + for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } + globaldata->Groups.clear(); + return 0; + } + + //get pscore for users tree + 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 - cout << "Tree " << i+1 << " parsimony score = " << userData[0] << endl; - //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]; + userData = pars->getValues(T[i]); //data = AB, AC, BC, ABC. - //save score for summary file - userTreeScores.push_back(userData[0]); - + if (m->control_pressed) { + delete reading; delete pars; delete util; delete output; + if (randomtree == "") { outSum.close(); } + for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } + globaldata->Groups.clear(); + return 0; + } + + + //output scores for each combination + for(int k = 0; k < numComp; k++) { + + //update uscoreFreq + map::iterator 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); - - //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; + + if (m->control_pressed) { + delete reading; delete pars; delete util; delete output; delete randT; + if (randomtree == "") { outSum.close(); } + for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } + globaldata->Groups.clear(); + return 0; } + + for(int r = 0; r < numComp; r++) { + //add trees pscore to map of scores + map::iterator it = rscoreFreq[r].find(randomData[r]); + if (it != 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); - //output info to pdistrib file - outDist << j+1 << '\t'<< '\t' << randomData[0] << endl; - 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(); + + if (m->control_pressed) { + delete reading; delete pars; delete util; delete output; delete randT; + globaldata->gTreemap = savetmap; + for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } + globaldata->Groups.clear(); + return 0; + } + + //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; + + if (m->control_pressed) { + delete reading; delete pars; delete util; delete output; delete randT; + globaldata->gTreemap = savetmap; + for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } + globaldata->Groups.clear(); + return 0; } - //add randoms score to validscores - validScores[randomData[0]] = randomData[0]; - + for(int r = 0; r < numComp; r++) { + //add trees pscore to map of scores + map::iterator it = rscoreFreq[r].find(randomData[r]); + if (it != 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[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 (map::iterator it = validScores.begin(); it != validScores.end(); it++) { + if (randomtree == "") { + map::iterator 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 + map::iterator 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]]); + if (m->control_pressed) { + delete reading; delete pars; delete util; delete output; + if (randomtree == "") { outSum.close(); } + else { globaldata->gTreemap = savetmap; } + for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } + globaldata->Groups.clear(); + return 0; } + + //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(); + globaldata->Groups.clear(); + + if (m->control_pressed) { + delete pars; delete util; delete output; + for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } + return 0; + } + + m->mothurOutEndLine(); + m->mothurOut("Output File Names: "); m->mothurOutEndLine(); + for (int i = 0; i < outputNames.size(); i++) { m->mothurOut(outputNames[i]); m->mothurOutEndLine(); } + m->mothurOutEndLine(); + return 0; } catch(exception& e) { - cout << "Standard Error: " << e.what() << " has occurred in the ParsimonyCommand class Function execute. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; - exit(1); - } - catch(...) { - cout << "An unknown error has occurred in the ParsimonyCommand class function execute. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + m->errorOut(e, "ParsimonyCommand", "execute"); exit(1); } } @@ -208,59 +349,67 @@ 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 (map::iterator 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"; - exit(1); - } - catch(...) { - cout << "An unknown error has occurred in the ParsimonyCommand class function printParsimonyFile. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + m->errorOut(e, "ParsimonyCommand", "printParsimonyFile"); exit(1); } } /***********************************************************/ -void ParsimonyCommand::printUSummaryFile() { +int ParsimonyCommand::printUSummaryFile() { try { //column headers - outSum << "Tree#" << '\t' << "ParsScore" << '\t' << '\t' << "ParsSig" << endl; + outSum << "Tree#" << '\t' << "Groups" << '\t' << "ParsScore" << '\t' << "ParsSig" << endl; + m->mothurOut("Tree#\tGroups\tParsScore\tParsSig"); m->mothurOutEndLine(); //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 (m->control_pressed) { outSum.close(); return 0; } + if (UScoreSig[a][i] > (1/(float)iters)) { + outSum << setprecision(6) << i+1 << '\t' << groupComb[a] << '\t' << userTreeScores[a][i] << setprecision(itersString.length()) << '\t' << UScoreSig[a][i] << endl; + cout << setprecision(6) << i+1 << '\t' << groupComb[a] << '\t' << userTreeScores[a][i] << setprecision(itersString.length()) << '\t' << UScoreSig[a][i] << endl; + m->mothurOutJustToLog(toString(i+1) + "\t" + groupComb[a] + "\t" + toString(userTreeScores[a][i]) + "\t" + toString(UScoreSig[a][i])); m->mothurOutEndLine(); + }else { + outSum << setprecision(6) << i+1 << '\t' << groupComb[a] << '\t' << userTreeScores[a][i] << setprecision(itersString.length()) << '\t' << "<" << (1/float(iters)) << endl; + cout << setprecision(6) << i+1 << '\t' << groupComb[a] << '\t' << userTreeScores[a][i] << setprecision(itersString.length()) << '\t' << "<" << (1/float(iters)) << endl; + m->mothurOutJustToLog(toString(i+1) + "\t" + groupComb[a] + "\t" + toString(userTreeScores[a][i]) + "\t" + toString((1/float(iters)))); m->mothurOutEndLine(); + } + } } outSum.close(); + return 0; } catch(exception& e) { - cout << "Standard Error: " << e.what() << " has occurred in the ParsimonyCommand class Function printUSummaryFile. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; - exit(1); - } - catch(...) { - cout << "An unknown error has occurred in the ParsimonyCommand class function printUSummaryFile. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + m->errorOut(e, "ParsimonyCommand", "printUSummaryFile"); exit(1); } } @@ -272,19 +421,22 @@ void ParsimonyCommand::getUserInput() { //create treemap tmap = new TreeMap(); - cout << "Please enter the number of groups you would like to analyze: "; + m->mothurOut("Please enter the number of groups you would like to analyze: "); cin >> numGroups; - + m->mothurOutJustToLog(toString(numGroups)); m->mothurOutEndLine(); + int num, count; count = 1; numEachGroup.resize(numGroups, 0); for (int i = 1; i <= numGroups; i++) { - cout << "Please enter the number of sequences in group " << i << ": "; + m->mothurOut("Please enter the number of sequences in group " + toString(i) + ": "); cin >> num; + m->mothurOutJustToLog(toString(num)); m->mothurOutEndLine(); //set tmaps seqsPerGroup tmap->seqsPerGroup[toString(i)] = num; + tmap->namesOfGroups.push_back(toString(i)); //set tmaps namesOfSeqs for (int j = 0; j < num; j++) { @@ -299,17 +451,18 @@ void ParsimonyCommand::getUserInput() { getline(cin, s); //save tmap for later + //memory leak prevention + //if (globaldata->gTreemap != NULL) { delete globaldata->gTreemap; } globaldata->gTreemap = tmap; + globaldata->Treenames = tmap->namesOfSeqs; } catch(exception& e) { - cout << "Standard Error: " << e.what() << " has occurred in the ParsimonyCommand class Function getUserInput. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; - exit(1); - } - catch(...) { - cout << "An unknown error has occurred in the ParsimonyCommand class function getUserInput. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + m->errorOut(e, "ParsimonyCommand", "getUserInput"); exit(1); } } + /***********************************************************/ +