]> git.donarmstrong.com Git - mothur.git/blobdiff - unifracweightedcommand.cpp
added bootstrap.shared command and fixed some bugs with heatmap
[mothur.git] / unifracweightedcommand.cpp
index 91e898f658556d840aab261995ba285c96d84f9d..6dfbd710765478d3e7b5c8dc21ab844fa4a6b231 100644 (file)
@@ -19,7 +19,12 @@ UnifracWeightedCommand::UnifracWeightedCommand() {
                sumFile = globaldata->getTreeFile() + ".wsummary";
                openOutputFile(sumFile, outSum);
                                
-               setGroups();    //sets the groups the user wants to analyze                     
+               util = new SharedUtil();
+               string s; //to make work with setgroups
+               util->setGroups(globaldata->Groups, tmap->namesOfGroups, s, numGroups, "weighted");     //sets the groups the user wants to analyze
+               util->getCombos(groupComb, globaldata->Groups, numComp);
+               globaldata->setGroups("");
+                               
                convert(globaldata->getIters(), iters);  //how many random trees to generate
                weighted = new Weighted(tmap);
 
@@ -36,6 +41,8 @@ UnifracWeightedCommand::UnifracWeightedCommand() {
 /***********************************************************/
 int UnifracWeightedCommand::execute() {
        try {
+               Progress* reading;
+               reading = new Progress("Comparing to random:", iters);
                
                //get weighted for users tree
                userData.resize(numComp,0);  //data[0] = weightedscore AB, data[1] = weightedscore AC...
@@ -49,10 +56,9 @@ int UnifracWeightedCommand::execute() {
                        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";
-                                                       
-                       cout << "Processing tree " << i+1 << endl;
+                       
+                       output = new ColumnFile(globaldata->getTreeFile()  + toString(i+1) + ".weighted");
+
                        userData = weighted->getValues(T[i]);  //userData[0] = weightedscore
                        
                        //save users score
@@ -82,6 +88,10 @@ int UnifracWeightedCommand::execute() {
                                                count++;
                                        }
                                }
+                               
+                               //update progress bar
+                               reading->update(j);
+
                        }
 
                        //removeValidScoresDuplicates(); 
@@ -104,12 +114,18 @@ int UnifracWeightedCommand::execute() {
                        calculateFreqsCumuls();
                        printWeightedFile();
                        
+                       delete output;
+                       
                        //clear data
                        rScores.clear();
                        uScores.clear();
                        validScores.clear();
                }
                
+               //finish progress bar
+               reading->finish();
+               delete reading;
+               
                printWSummaryFile();
                
                //clear out users groups
@@ -133,22 +149,19 @@ int UnifracWeightedCommand::execute() {
 void UnifracWeightedCommand::printWeightedFile() {
        try {
                vector<double> data;
+               vector<string> tags;
+               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++) { 
                                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(weightedFileout.c_str());
-               
        }
        catch(exception& e) {
                cout << "Standard Error: " << e.what() << " has occurred in the UnifracWeightedCommand class Function printWeightedFile. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
@@ -165,8 +178,8 @@ void UnifracWeightedCommand::printWeightedFile() {
 void UnifracWeightedCommand::printWSummaryFile() {
        try {
                //column headers
-               outSum << "Tree#" << '\t' << "Groups" << '\t' << '\t' << "WScore" << '\t' << '\t' << "WSig" <<  endl;
-               cout << "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);
@@ -175,8 +188,13 @@ void UnifracWeightedCommand::printWSummaryFile() {
                int count = 0;
                for (int i = 0; i < T.size(); i++) { 
                        for (int j = 0; j < numComp; j++) {
-                               outSum << setprecision(globaldata->getIters().length()) << i+1 << '\t' << '\t' << groupComb[j] << '\t' << utreeScores[count] << '\t' << WScoreSig[count] << endl; 
-                               cout << setprecision(globaldata->getIters().length()) << 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' << groupComb[j] << '\t' << utreeScores[count] << '\t' << setprecision(globaldata->getIters().length()) << WScoreSig[count] << endl; 
+                                       cout << setprecision(6) << i+1 << '\t' << groupComb[j] << '\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++;
                        }
                }
@@ -210,71 +228,6 @@ int UnifracWeightedCommand::findIndex(float score, int index) {
        }
 }
 
-/***********************************************************/
-void UnifracWeightedCommand::setGroups() {
-       try {
-               //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; i<numGroups; i++) { 
-                       numComp += i; 
-                       for (int l = i+1; l < numGroups; l++) {
-                               //set group comparison labels
-                               groupComb.push_back(globaldata->Groups[i]+globaldata->Groups[l]);
-                       }
-               }
-       }
-       catch(exception& e) {
-               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 setGroups. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
-               exit(1);
-       }
-}
-
 /***********************************************************/
 
 void UnifracWeightedCommand::calculateFreqsCumuls() {
@@ -325,84 +278,9 @@ void UnifracWeightedCommand::calculateFreqsCumuls() {
 
 }
 
-/*****************************************************************/
-
-void UnifracWeightedCommand::initFile(string label){
-       try {
-               if(counter != 0){
-                       openOutputFile(weightedFileout, out);
-                       openInputFile(weightedFile, inFile);
-
-                       string inputBuffer;
-                       getline(inFile, inputBuffer);
-               
-                       out     <<  inputBuffer << '\t' << label + "Score" << '\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<double> data){
-       try {
-               if(counter != 0){               
-                       string inputBuffer;
-                       getline(inFile, inputBuffer);
-               
-                       out     <<  inputBuffer << setprecision(globaldata->getIters().length()) << '\t' << data[0] << '\t' << data[1] << '\t' << data[2] << endl;
-               }
-               else{
-                       out << setprecision(globaldata->getIters().length()) << data[0] << '\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);
-       }       
-}