X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=parsimonycommand.cpp;h=266dd3b9a2ad5c3e7730ac6950e7077164b12bb2;hb=786d5631d9cd5baa6ed6ef16f8b4b384cbc7470f;hp=c40dde0c86c323560c850348940530293a1f6659;hpb=cfeefe1f7d36e9ffd3fba5f2f2b906724393e442;p=mothur.git diff --git a/parsimonycommand.cpp b/parsimonycommand.cpp index c40dde0..266dd3b 100644 --- a/parsimonycommand.cpp +++ b/parsimonycommand.cpp @@ -10,32 +10,78 @@ #include "parsimonycommand.h" /***********************************************************/ -ParsimonyCommand::ParsimonyCommand() { +ParsimonyCommand::ParsimonyCommand(string option) { try { globaldata = GlobalData::getInstance(); + abort = false; + Groups.clear(); - //randomtree will tell us if user had their own treefile or if they just want the random distribution - randomtree = globaldata->getRandomTree(); + //allow user to run help + if(option == "help") { help(); abort = true; } - //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); - }else { //user wants random distribution - savetmap = globaldata->gTreemap; - getUserInput(); - parsFile = randomtree + ".rd_parsimony"; - openOutputFile(parsFile, out); - } + else { + //valid paramters for this command + string Array[] = {"random","groups","iters"}; + vector myArray (Array, Array+(sizeof(Array)/sizeof(string))); + + OptionParser parser(option); + map parameters = parser.getParameters(); + + ValidParameters validParameter; - //set users groups to analyze - setGroups(); - convert(globaldata->getIters(), iters); //how many random trees to generate - pars = new Parsimony(tmap); + //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; } + } + + randomtree = validParameter.validFile(parameters, "random", false); if (randomtree == "not found") { randomtree = ""; } + + //are you trying to use parsimony without reading a tree or saying you want random distribution + if (randomtree == "") { + if (globaldata->gTree.size() == 0) { + cout << "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." << endl; abort = true; } + } + + //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 = ""; } + 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; + output = new ColumnFile(globaldata->getTreeFile() + ".parsimony", itersString); + sumFile = globaldata->getTreeFile() + ".psummary"; + openOutputFile(sumFile, outSum); + }else { //user wants random distribution + savetmap = globaldata->gTreemap; + getUserInput(); + output = new ColumnFile(randomtree, itersString); + } + + //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); + + if (numGroups == 1) { numComp++; groupComb.push_back(allGroups); } + + pars = new Parsimony(tmap); + counter = 0; + + } + + } } catch(exception& e) { @@ -47,10 +93,41 @@ ParsimonyCommand::ParsimonyCommand() { exit(1); } } + +//********************************************************************************************************************** + +void ParsimonyCommand::help(){ + try { + cout << "The parsimony command can only be executed after a successful read.tree command, unless you use the random parameter." << "\n"; + cout << "The parsimony command parameters are random, groups and iters. No parameters are required." << "\n"; + cout << "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"; + cout << "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"; + cout << "The parsimony command should be in the following format: parsimony(random=yourOutputFilename, groups=yourGroups, iters=yourIters)." << "\n"; + cout << "Example parsimony(random=out, iters=500)." << "\n"; + cout << "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"; + cout << "and iters is 1000. The parsimony command output two files: .parsimony and .psummary their descriptions are in the manual." << "\n"; + cout << "Note: No spaces between parameter labels (i.e. random), '=' and parameters (i.e.yourOutputFilename)." << "\n" << "\n"; + } + catch(exception& e) { + cout << "Standard Error: " << e.what() << " has occurred in the ParsimonyCommand class Function help. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + exit(1); + } + catch(...) { + cout << "An unknown error has occurred in the ParsimonyCommand class function help. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + 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(numComp,0); //data = AB, AC, BC, ABC. randomData.resize(numComp,0); //data = AB, AC, BC, ABC. @@ -58,27 +135,25 @@ int ParsimonyCommand::execute() { uscoreFreq.resize(numComp); rCumul.resize(numComp); uCumul.resize(numComp); - validScores.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]); //data = AB, AC, BC, ABC. - + //output scores for each combination for(int k = 0; k < numComp; k++) { - cout << "Tree " << i+1 << " Combination " << groupComb[k] << " parsimony score = " << userData[k] << endl; + //update uscoreFreq - it = uscoreFreq[k].find(userData[k]); + 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[k][userData[k]] = userData[k]; + validScores[userData[k]] = userData[k]; //save score for summary file userTreeScores[k].push_back(userData[k]); @@ -89,74 +164,84 @@ int ParsimonyCommand::execute() { 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 + 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[r][randomData[r]] = randomData[r]; + validScores[randomData[r]] = randomData[r]; } + //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); - + 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 + 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[r][randomData[r]] = randomData[r]; + validScores[randomData[r]] = randomData[r]; } + //update progress bar + reading->update(j); + delete randT; } } - - float rcumul = 0.0000; - float ucumul = 0.0000; - + for(int a = 0; a < numComp; a++) { - //this loop fills the cumulative maps and put 0.0000 in the score freq map to make it easier to print. - for (it = validScores[a].begin(); it != validScores[a].end(); it++) { + 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 == "") { - it2 = uscoreFreq[a].find(it->first); + 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-a; + uCumul[a][it->first] = ucumul; } //make rscoreFreq map and rCumul - it2 = rscoreFreq[a].find(it->first); + 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-a; + rCumul[a][it->first] = rcumul; } //find the signifigance of each user trees score when compared to the random trees and save for printing the summary file @@ -165,16 +250,23 @@ int ParsimonyCommand::execute() { } } + //finish progress bar + reading->finish(); + delete reading; + + printParsimonyFile(); - if (randomtree != "") { 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->setGroups(""); + globaldata->Groups.clear(); return 0; @@ -192,28 +284,29 @@ int ParsimonyCommand::execute() { /***********************************************************/ void ParsimonyCommand::printParsimonyFile() { try { - //column headers + vector data; + vector tags; + if (randomtree == "") { - out << "Comb" << '\t' << "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 << "Comb" << '\t' << "Score" << '\t' << "RandFreq" << '\t' << "RandCumul" << endl; + tags.push_back("Score"); tags.push_back("RandFreq"); tags.push_back("RandCumul"); } - - //format output - out.setf(ios::fixed, ios::floatfield); out.setf(ios::showpoint); - + for(int a = 0; a < numComp; a++) { + output->initFile(groupComb[a], tags); //print each line - for (it = validScores[a].begin(); it != validScores[a].end(); it++) { + for (map::iterator it = validScores.begin(); it != validScores.end(); it++) { if (randomtree == "") { - 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; + 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{ - out << setprecision(6) << groupComb[a] << '\t' << it->first << '\t' << '\t' << rscoreFreq[a][it->first] << '\t' << rCumul[a][it->first] << endl; + 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(); } - 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"; @@ -228,7 +321,8 @@ void ParsimonyCommand::printParsimonyFile() { void ParsimonyCommand::printUSummaryFile() { try { //column headers - outSum << "Tree#" << '\t' << "Comb" << '\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); @@ -237,7 +331,13 @@ void ParsimonyCommand::printUSummaryFile() { //print each line for (int i = 0; i< T.size(); i++) { for(int a = 0; a < numComp; a++) { - outSum << setprecision(6) << i+1 << '\t' << groupComb[a] << '\t' << '\t' << userTreeScores[a][i] << '\t' << UScoreSig[a][i] << endl; + 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; + }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; + } } } @@ -288,7 +388,10 @@ 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) { @@ -300,77 +403,7 @@ void ParsimonyCommand::getUserInput() { exit(1); } } -/***********************************************************/ - -void ParsimonyCommand::setGroups() { - try { - string allGroups = ""; - numGroups = 0; - //if the user has not entered specific groups to analyze then do them all - if (globaldata->Groups.size() != 0) { - if (globaldata->Groups[0] != "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) { - 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]); - numGroups++; - allGroups += tmap->namesOfGroups[i]; - } - }else { - for (int i = 0; i < globaldata->Groups.size(); i++) { - allGroups += tmap->namesOfGroups[i]; - numGroups++; - } - } - }else{//user has enter "all" and wants the default groups - for (int i = 0; i < tmap->namesOfGroups.size(); i++) { - globaldata->Groups.push_back(tmap->namesOfGroups[i]); - numGroups++; - allGroups += tmap->namesOfGroups[i]; - } - globaldata->setGroups(""); - } - }else { - for (int i = 0; i < tmap->namesOfGroups.size(); i++) { - allGroups += tmap->namesOfGroups[i]; - } - numGroups = 1; - } - - //calculate number of comparsions - numComp = 0; - for (int r=0; rGroups[r]+globaldata->Groups[l]); - numComp++; - } - } - - //ABC - groupComb.push_back(allGroups); - numComp++; - - } - catch(exception& e) { - cout << "Standard Error: " << e.what() << " has occurred in the ParsimonyCommand class Function setGroups. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; - exit(1); - } - catch(...) { - cout << "An unknown error has occurred in the ParsimonyCommand class function setGroups. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; - exit(1); - } -} -/*****************************************************************/ +/***********************************************************/