]> git.donarmstrong.com Git - mothur.git/blobdiff - parsimonycommand.cpp
added alignment code
[mothur.git] / parsimonycommand.cpp
index 4529f777358ece0b03a6f53d522a54bd7698ac87..56305e2260d4af19050dfc0832add36c1208d13e 100644 (file)
@@ -21,19 +21,23 @@ ParsimonyCommand::ParsimonyCommand() {
                if (randomtree == "") { 
                        T = globaldata->gTree;
                        tmap = globaldata->gTreemap;
-                       parsFile = globaldata->getTreeFile() + ".parsimony";
-                       parsFileout = globaldata->getTreeFile() + "temp" + ".parsimony";
+                       output = new ColumnFile(globaldata->getTreeFile()  +  ".parsimony");
                        sumFile = globaldata->getTreeFile() + ".psummary";
                        openOutputFile(sumFile, outSum);
                }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
-               setGroups();
+               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;
@@ -51,7 +55,9 @@ ParsimonyCommand::ParsimonyCommand() {
 /***********************************************************/
 int ParsimonyCommand::execute() {
        try {
-       
+               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.
@@ -65,11 +71,11 @@ int ParsimonyCommand::execute() {
                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++) {
+
                                        //update uscoreFreq
                                        it = uscoreFreq[k].find(userData[k]);
                                        if (it == uscoreFreq[k].end()) {//new score
@@ -88,11 +94,13 @@ 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]);
@@ -106,18 +114,24 @@ int ParsimonyCommand::execute() {
                                        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]);
@@ -131,10 +145,13 @@ int ParsimonyCommand::execute() {
                                        validScores[randomData[r]] = randomData[r];
                                }
                                
+                               //update progress bar
+                               reading->update(j);
+                               
                                delete randT;
                        }
                }
-               
+
                for(int a = 0; a < numComp; a++) {
                        float rcumul = 0.0000;
                        float ucumul = 0.0000;
@@ -163,16 +180,25 @@ int ParsimonyCommand::execute() {
                        }
                }
                
+               //finish progress bar
+               reading->finish();
+               delete reading;
+
+               
                printParsimonyFile();
                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;
                
@@ -191,12 +217,16 @@ int ParsimonyCommand::execute() {
 void ParsimonyCommand::printParsimonyFile() {
        try {
                vector<double> data;
+               vector<string> tags;
                
-               //format output
-               out.setf(ios::fixed, ios::floatfield); out.setf(ios::showpoint);
+               if (randomtree == "") {
+                       tags.push_back("Score"); tags.push_back("UserFreq"); tags.push_back("UserCumul"); tags.push_back("RandFreq"); tags.push_back("RandCumul");
+               }else {
+                       tags.push_back("Score"); tags.push_back("RandFreq"); tags.push_back("RandCumul");
+               }
 
                for(int a = 0; a < numComp; a++) {
-                       initFile(groupComb[a]);
+                       output->initFile(groupComb[a], tags);
                        //print each line
                        for (it = validScores.begin(); it != validScores.end(); it++) { 
                                if (randomtree == "") {
@@ -204,15 +234,11 @@ void ParsimonyCommand::printParsimonyFile() {
                                }else{
                                        data.push_back(it->first);  data.push_back(rscoreFreq[a][it->first]); data.push_back(rCumul[a][it->first]); 
                                }
-                               output(data);
+                               output->output(data);
                                data.clear();
                        } 
-                       resetFile();
+                       output->resetFile();
                }
-               
-               out.close();
-               inFile.close();
-               remove(parsFileout.c_str());
        }
        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";
@@ -227,7 +253,8 @@ 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);
@@ -236,8 +263,13 @@ 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;
-                               cout << 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(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;
+                               }
                        }
                }
                
@@ -288,6 +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;
                
        }
@@ -300,175 +334,7 @@ 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; r<numGroups; r++) { 
-                       for (int l = r+1; l < numGroups; l++) {
-                               groupComb.push_back(globaldata->Groups[r]+globaldata->Groups[l]);
-                               numComp++;
-                       }
-               }
-               
-               //ABC
-               if (numComp != 1) {
-                       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);
-       }               
-
-}
-/*****************************************************************/
-
-void ParsimonyCommand::initFile(string label){
-       try {
-               if(counter != 0){
-                       openOutputFile(parsFileout, out);
-                       openInputFile(parsFile, inFile);
-
-                       string inputBuffer;
-                       getline(inFile, inputBuffer);
-                       
-                       if (randomtree == "") {
-                               out <<  inputBuffer << '\t' << label + "Score" << '\t' << label + "UserFreq" << '\t' << label + "UserCumul" << '\t' << label + "RandFreq" << '\t' << label + "RandCumul" << endl;
-                       }else {
-                               out <<  inputBuffer << '\t' << label + "Score" << '\t' << label + "RandFreq" << '\t' << label + "RandCumul" << endl;
-                       }
-               }else{
-                       openOutputFile(parsFileout, out);
-                       //column headers
-                       if (randomtree == "") {
-                               out << label + "Score" << '\t' << label + "UserFreq" << '\t' << label + "UserCumul" << '\t' << label + "RandFreq" << '\t' << label + "RandCumul" << endl;
-                       }else {
-                               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 ParsimonyCommand class Function initFile. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
-               exit(1);
-       }
-       catch(...) {
-               cout << "An unknown error has occurred in the ParsimonyCommand class function initFile. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
-               exit(1);
-       }
-}
-
-/***********************************************************************/
-
-void ParsimonyCommand::output(vector<double> data){
-       try {
-               if(counter != 0){               
-                       string inputBuffer;
-                       getline(inFile, inputBuffer);
-               
-                       if (randomtree == "") {
-                               out << inputBuffer << '\t' << setprecision(6) << data[0] << '\t' << data[1] << '\t' << data[2] << '\t' << data[3] << '\t' << data[4] << endl;
-                       }else{
-                               out << inputBuffer << '\t' << setprecision(6) << data[0] << '\t' << data[1] << '\t' << data[2] << endl;
-                       }
-               }
-               else{
-                       if (randomtree == "") {
-                               out << setprecision(6) << data[0] << '\t' << data[1] << '\t' << data[2] << '\t' << data[3] << '\t' << data[4] << endl;
-                       }else{
-                               out << setprecision(6) << data[0] << '\t' << data[1] << '\t' << data[2] << endl;
-                       }
-               }
-
-       }
-       catch(exception& e) {
-               cout << "Standard Error: " << e.what() << " has occurred in the ParsimonyCommand class Function output. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
-               exit(1);
-       }
-       catch(...) {
-               cout << "An unknown error has occurred in the ParsimonyCommand class function output. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
-               exit(1);
-       }
-}
-
-/***********************************************************************/
-
-void ParsimonyCommand::resetFile(){
-       try {
-               if(counter != 0){
-                       out.close();
-                       inFile.close();
-               }
-               else{
-                       out.close();
-               }
-               counter = 1;
-               
-               remove(parsFile.c_str());
-               rename(parsFileout.c_str(), parsFile.c_str());
-       }
-       catch(exception& e) {
-               cout << "Standard Error: " << e.what() << " has occurred in the ParsimonyCommand class Function resetFile. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
-               exit(1);
-       }
-       catch(...) {
-               cout << "An unknown error has occurred in the ParsimonyCommand class function resetFile. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
-               exit(1);
-       }       
-}
+/***********************************************************/