]> git.donarmstrong.com Git - mothur.git/blobdiff - parsimonycommand.cpp
Merge remote-tracking branch 'origin/master'
[mothur.git] / parsimonycommand.cpp
index 66f8193e647c94db5bc8ac815bb6eea9796cd138..f124b6002ee2c247db14716b4be81cb9ec10d8f2 100644 (file)
@@ -8,6 +8,7 @@
  */
 
 #include "parsimonycommand.h"
+#include "treereader.h"
 
 //**********************************************************************************************************************
 vector<string> ParsimonyCommand::setParameters(){      
@@ -51,6 +52,27 @@ string ParsimonyCommand::getHelpString(){
                exit(1);
        }
 }
+//**********************************************************************************************************************
+string ParsimonyCommand::getOutputFileNameTag(string type, string inputName=""){       
+       try {
+        string outputFileName = "";
+               map<string, vector<string> >::iterator it;
+        
+        //is this a type this command creates
+        it = outputTypes.find(type);
+        if (it == outputTypes.end()) {  m->mothurOut("[ERROR]: this command doesn't create a " + type + " output file.\n"); }
+        else {
+            if (type == "parsimony") {  outputFileName =  "parsimony"; }
+            else if (type == "psummary") {  outputFileName =  "psummary"; }
+            else { m->mothurOut("[ERROR]: No definition for type " + type + " output file tag.\n"); m->control_pressed = true;  }
+        }
+        return outputFileName;
+       }
+       catch(exception& e) {
+               m->errorOut(e, "ParsimonyCommand", "getOutputFileNameTag");
+               exit(1);
+       }
+}
 
 //**********************************************************************************************************************
 ParsimonyCommand::ParsimonyCommand(){  
@@ -125,12 +147,6 @@ ParsimonyCommand::ParsimonyCommand(string option)  {
                                }
                        }
                        
-                       m->runParse = true;
-                       m->Groups.clear();
-                       m->namesOfGroups.clear();
-                       m->Treenames.clear();
-                       m->names.clear();
-                       
                        outputDir = validParameter.validFile(parameters, "outputdir", false);           if (outputDir == "not found"){  outputDir = ""; }
                        
                        randomtree = validParameter.validFile(parameters, "random", false);             if (randomtree == "not found") { randomtree = ""; }
@@ -139,7 +155,7 @@ ParsimonyCommand::ParsimonyCommand(string option)  {
                        if (randomtree == "")  {
                                //check for required parameters
                                treefile = validParameter.validFile(parameters, "tree", true);
-                               if (treefile == "not open") { abort = true; }
+                               if (treefile == "not open") { treefile = ""; abort = true; }
                                else if (treefile == "not found") {                             //if there is a current design file, use it
                                        treefile = m->getTreeFile(); 
                                        if (treefile != "") { m->mothurOut("Using " + treefile + " as input file for the tree parameter."); m->mothurOutEndLine(); }
@@ -153,7 +169,7 @@ ParsimonyCommand::ParsimonyCommand(string option)  {
                                else { m->setGroupFile(groupfile); }
                                
                                namefile = validParameter.validFile(parameters, "name", true);
-                               if (namefile == "not open") { abort = true; }
+                               if (namefile == "not open") { namefile = ""; abort = true; }
                                else if (namefile == "not found") { namefile = ""; }
                                else { m->setNameFile(namefile); }
                        }
@@ -164,18 +180,23 @@ ParsimonyCommand::ParsimonyCommand(string option)  {
                        //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 = ""; m->Groups.clear(); }
+                       if (groups == "not found") { groups = ""; m->clearGroups(); }
                        else { 
                                m->splitAtDash(groups, Groups);
-                               m->Groups = Groups;
+                               m->setGroups(Groups);
                        }
                                
                        itersString = validParameter.validFile(parameters, "iters", false);                     if (itersString == "not found") { itersString = "1000"; }
-                       convert(itersString, iters); 
+                       m->mothurConvert(itersString, iters); 
                        
                        string temp = validParameter.validFile(parameters, "processors", false);        if (temp == "not found"){       temp = m->getProcessors();      }
                        m->setProcessors(temp);
-                       convert(temp, processors);
+                       m->mothurConvert(temp, processors);
+                       
+                       if (namefile == "") {
+                               vector<string> files; files.push_back(treefile);
+                               parser.getNameFile(files);
+                       }
                        
                }
 
@@ -198,74 +219,17 @@ int ParsimonyCommand::execute() {
                        
                        m->setTreeFile(treefile);
                        
-                       if (groupfile != "") {
-                               //read in group map info.
-                               tmap = new TreeMap(groupfile);
-                               tmap->readMap();
-                       }else{ //fake out by putting everyone in one group
-                               Tree* tree = new Tree(treefile); delete tree;  //extracts names from tree to make faked out groupmap
-                               tmap = new TreeMap();
-                               
-                               for (int i = 0; i < m->Treenames.size(); i++) { tmap->addSeq(m->Treenames[i], "Group1"); }
-                       }
-                       
-                       if (namefile != "") { readNamesFile(); }
-                       
-                       read = new ReadNewickTree(treefile);
-                       int readOk = read->read(tmap); 
-                       
-                       if (readOk != 0) { m->mothurOut("Read Terminated."); m->mothurOutEndLine(); delete tmap; delete read; return 0; }
-                       
-                       read->AssembleTrees();
-                       T = read->getTrees();
-                       delete read;
-
-                       //make sure all files match
-                       //if you provide a namefile we will use the numNames in the namefile as long as the number of unique match the tree names size.
-                       int numNamesInTree;
-                       if (namefile != "")  {  
-                               if (numUniquesInName == m->Treenames.size()) {  numNamesInTree = nameMap.size();  }
-                               else {   numNamesInTree = m->Treenames.size();  }
-                       }else {  numNamesInTree = m->Treenames.size();  }
-                       
-                       
-                       //output any names that are in group file but not in tree
-                       if (numNamesInTree < tmap->getNumSeqs()) {
-                               for (int i = 0; i < tmap->namesOfSeqs.size(); i++) {
-                                       //is that name in the tree?
-                                       int count = 0;
-                                       for (int j = 0; j < m->Treenames.size(); j++) {
-                                               if (tmap->namesOfSeqs[i] == m->Treenames[j]) { break; } //found it
-                                               count++;
-                                       }
-                                       
-                                       if (m->control_pressed) { 
-                                               delete tmap; for (int i = 0; i < T.size(); i++) { delete T[i]; }
-                                               for (int i = 0; i < outputNames.size(); i++) {  m->mothurRemove(outputNames[i]); } outputTypes.clear();
-                                               m->Groups.clear();
-                                               return 0;
-                                       }
-                                       
-                                       //then you did not find it so report it 
-                                       if (count == m->Treenames.size()) { 
-                                               //if it is in your namefile then don't remove
-                                               map<string, string>::iterator it = nameMap.find(tmap->namesOfSeqs[i]);
-                                               
-                                               if (it == nameMap.end()) {
-                                                       m->mothurOut(tmap->namesOfSeqs[i] + " is in your groupfile and not in your tree. It will be disregarded."); m->mothurOutEndLine();
-                                                       tmap->removeSeq(tmap->namesOfSeqs[i]);
-                                                       i--; //need this because removeSeq removes name from namesOfSeqs
-                                               }
-                                       }
-                               }
-                       }
-                               
+            TreeReader* reader = new TreeReader(treefile, groupfile, namefile);
+            T = reader->getTrees();
+            tmap = T[0]->getTreeMap();
+            delete reader;
+       
                        if(outputDir == "") { outputDir += m->hasPath(treefile); }
-                       output = new ColumnFile(outputDir + m->getSimpleName(treefile)  +  ".parsimony", itersString);
-                       outputNames.push_back(outputDir + m->getSimpleName(treefile)  +  ".parsimony");
-                       outputTypes["parsimony"].push_back(outputDir + m->getSimpleName(treefile)  +  ".parsimony");
+                       output = new ColumnFile(outputDir + m->getSimpleName(treefile)  +  "." + getOutputFileNameTag("parsimony"), itersString);
+                       outputNames.push_back(outputDir + m->getSimpleName(treefile)  +  "." + getOutputFileNameTag("parsimony"));
+                       outputTypes["parsimony"].push_back(outputDir + m->getSimpleName(treefile)  + "." + getOutputFileNameTag("parsimony"));
                                
-                       sumFile = outputDir + m->getSimpleName(treefile) + ".psummary";
+                       sumFile = outputDir + m->getSimpleName(treefile) + "." + getOutputFileNameTag("psummary");
                        m->openOutputFile(sumFile, outSum);
                        outputNames.push_back(sumFile);
                        outputTypes["psummary"].push_back(sumFile);
@@ -279,25 +243,27 @@ int ParsimonyCommand::execute() {
                }
                        
                //set users groups to analyze
-               util = new SharedUtil();
-               util->setGroups(m->Groups, tmap->namesOfGroups, allGroups, numGroups, "parsimony");     //sets the groups the user wants to analyze
-               util->getCombos(groupComb, m->Groups, numComp);
-               delete util;
+               SharedUtil util;
+               vector<string> mGroups = m->getGroups();
+               vector<string> tGroups = tmap->getNamesOfGroups();
+               util.setGroups(mGroups, tGroups, allGroups, numGroups, "parsimony");    //sets the groups the user wants to analyze
+               util.getCombos(groupComb, mGroups, numComp);
+               m->setGroups(mGroups);
                        
                if (numGroups == 1) { numComp++; groupComb.push_back(allGroups); }
                        
-               pars = new Parsimony(tmap);
+               Parsimony pars;
                counter = 0;
        
                Progress* reading;
                reading = new Progress("Comparing to random:", iters);
                
                if (m->control_pressed) { 
-                       delete reading; delete pars; delete output;
+                       delete reading; delete output;
                        delete tmap; for (int i = 0; i < T.size(); i++) { delete T[i]; }
                        if (randomtree == "") {  outSum.close();  }
                        for (int i = 0; i < outputNames.size(); i++) {  m->mothurRemove(outputNames[i]); } outputTypes.clear();
-                       m->Groups.clear();
+                       m->clearGroups();
                        return 0;
                }
                        
@@ -315,14 +281,14 @@ int ParsimonyCommand::execute() {
                if (randomtree == "") {
                        //get pscores for users trees
                        for (int i = 0; i < T.size(); i++) {
-                               userData = pars->getValues(T[i], processors, outputDir);  //data = AB, AC, BC, ABC.
+                               userData = pars.getValues(T[i], processors, outputDir);  //data = AB, AC, BC, ABC.
                                
                                if (m->control_pressed) { 
-                                       delete reading; delete pars; delete output;
+                                       delete reading; delete output;
                                        delete tmap; for (int i = 0; i < T.size(); i++) { delete T[i]; }
                                        if (randomtree == "") {  outSum.close();  }
                                        for (int i = 0; i < outputNames.size(); i++) {  m->mothurRemove(outputNames[i]); } outputTypes.clear();
-                                       m->Groups.clear();
+                                       m->clearGroups();
                                        return 0;
                                }
 
@@ -354,14 +320,14 @@ int ParsimonyCommand::execute() {
                                randT->assembleRandomTree();
 
                                //get pscore of random tree
-                               randomData = pars->getValues(randT, processors, outputDir);
+                               randomData = pars.getValues(randT, processors, outputDir);
                                
                                if (m->control_pressed) { 
-                                       delete reading; delete pars; delete output; delete randT;
+                                       delete reading;  delete output; delete randT;
                                        if (randomtree == "") {  outSum.close();  }
                                        for (int i = 0; i < outputNames.size(); i++) {  m->mothurRemove(outputNames[i]); } outputTypes.clear();
                                        delete tmap; for (int i = 0; i < T.size(); i++) { delete T[i]; }
-                                       m->Groups.clear();
+                                       m->clearGroups();
                                        return 0;
                                }
                                        
@@ -395,23 +361,17 @@ int ParsimonyCommand::execute() {
                                randT->assembleRandomTree();
                                
                                if (m->control_pressed) { 
-                                       delete reading; delete pars; delete output; delete randT;
-                                       delete tmap; 
-                                       for (int i = 0; i < outputNames.size(); i++) {  m->mothurRemove(outputNames[i]); } outputTypes.clear();
-                                       m->Groups.clear();
-                                       return 0;
+                                       delete reading; delete output; delete randT; delete tmap; 
+                                       for (int i = 0; i < outputNames.size(); i++) {  m->mothurRemove(outputNames[i]); } outputTypes.clear(); return 0;
                                }
 
 
                                //get pscore of random tree
-                               randomData = pars->getValues(randT, processors, outputDir);
+                               randomData = pars.getValues(randT, processors, outputDir);
                                
                                if (m->control_pressed) { 
-                                       delete reading; delete pars;  delete output; delete randT;
-                                       delete tmap; 
-                                       for (int i = 0; i < outputNames.size(); i++) {  m->mothurRemove(outputNames[i]); } outputTypes.clear();
-                                       m->Groups.clear();
-                                       return 0;
+                                       delete reading; delete output; delete randT; delete tmap; 
+                                       for (int i = 0; i < outputNames.size(); i++) {  m->mothurRemove(outputNames[i]); } outputTypes.clear(); return 0;
                                }
                        
                                for(int r = 0; r < numComp; r++) {
@@ -463,27 +423,21 @@ int ParsimonyCommand::execute() {
                }
                
                if (m->control_pressed) { 
-                               delete reading; delete pars; delete output;
+                               delete reading; delete output;
                                delete tmap; for (int i = 0; i < T.size(); i++) { delete T[i]; }
                                if (randomtree == "") {  outSum.close();  }
                                for (int i = 0; i < outputNames.size(); i++) {  m->mothurRemove(outputNames[i]); } outputTypes.clear();
-                               m->Groups.clear();
                                return 0;
                }
                
                //finish progress bar
                reading->finish();
                delete reading;
-
                
                printParsimonyFile();
                if (randomtree == "") { printUSummaryFile(); }
-               
-               //reset groups parameter
-               m->Groups.clear(); 
-               
-               delete pars; delete output; 
-               delete tmap; for (int i = 0; i < T.size(); i++) { delete T[i]; }
+                               
+        delete output; delete tmap; for (int i = 0; i < T.size(); i++) { delete T[i]; }
                
                if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) {        m->mothurRemove(outputNames[i]); } outputTypes.clear(); return 0;}
                
@@ -585,6 +539,7 @@ void ParsimonyCommand::getUserInput() {
                count = 1;
                numEachGroup.resize(numGroups, 0);  
                
+               
                for (int i = 1; i <= numGroups; i++) {
                        m->mothurOut("Please enter the number of sequences in group " + toString(i) +  ": ");
                        cin >> num;
@@ -592,7 +547,7 @@ void ParsimonyCommand::getUserInput() {
                                
                        //set tmaps seqsPerGroup
                        tmap->seqsPerGroup[toString(i)] = num;
-                       tmap->namesOfGroups.push_back(toString(i));
+                       tmap->addGroup(toString(i));
                        
                        //set tmaps namesOfSeqs
                        for (int j = 0; j < num; j++) {
@@ -614,46 +569,6 @@ void ParsimonyCommand::getUserInput() {
                exit(1);
        }
 }
-/*****************************************************************/
-int ParsimonyCommand::readNamesFile() {
-       try {
-               m->names.clear();
-               numUniquesInName = 0;
-               
-               ifstream in;
-               m->openInputFile(namefile, in);
-               
-               string first, second;
-               map<string, string>::iterator itNames;
-               
-               while(!in.eof()) {
-                       in >> first >> second; m->gobble(in);
-                       
-                       numUniquesInName++;
-                       
-                       itNames = m->names.find(first);
-                       if (itNames == m->names.end()) {  
-                               m->names[first] = second; 
-                               
-                               //we need a list of names in your namefile to use above when removing extra seqs above so we don't remove them
-                               vector<string> dupNames;
-                               m->splitAtComma(second, dupNames);
-                               
-                               for (int i = 0; i < dupNames.size(); i++) {     
-                                       nameMap[dupNames[i]] = dupNames[i]; 
-                                       if ((groupfile == "") && (i != 0)) { tmap->addSeq(dupNames[i], "Group1"); } 
-                               }
-                       }else {  m->mothurOut(first + " has already been seen in namefile, disregarding names file."); m->mothurOutEndLine(); in.close(); m->names.clear(); namefile = ""; return 1; }                  
-               }
-               in.close();
-               
-               return 0;
-       }
-       catch(exception& e) {
-               m->errorOut(e, "ParsimonyCommand", "readNamesFile");
-               exit(1);
-       }
-}
 /***********************************************************/