]> git.donarmstrong.com Git - mothur.git/blobdiff - treegroupscommand.cpp
changed dups parameter to dereplicate in chimera.uchime.
[mothur.git] / treegroupscommand.cpp
index 888f50202e3443855a99d82d71b126cf715ee380..bba6289ada54856c25d683126b0d181badab4c15 100644 (file)
@@ -16,8 +16,9 @@ vector<string> TreeGroupCommand::setParameters(){
        try {
                CommandParameter pshared("shared", "InputTypes", "", "", "PhylipColumnShared", "PhylipColumnShared", "none",false,false); parameters.push_back(pshared);
                CommandParameter pphylip("phylip", "InputTypes", "", "", "PhylipColumnShared", "PhylipColumnShared", "none",false,false); parameters.push_back(pphylip);
-               CommandParameter pname("name", "InputTypes", "", "", "none", "none", "ColumnName",false,false); parameters.push_back(pname);
-               CommandParameter pcolumn("column", "InputTypes", "", "", "PhylipColumnShared", "PhylipColumnShared", "ColumnName",false,false); parameters.push_back(pcolumn);  
+               CommandParameter pname("name", "InputTypes", "", "", "NameCount", "none", "ColumnName",false,false); parameters.push_back(pname);
+               CommandParameter pcount("count", "InputTypes", "", "", "NameCount", "none", "countcolumn",false,false); parameters.push_back(pcount);
+        CommandParameter pcolumn("column", "InputTypes", "", "", "PhylipColumnShared", "PhylipColumnShared", "ColumnName-countcolumn",false,false); parameters.push_back(pcolumn);             
         CommandParameter piters("iters", "Number", "", "1000", "", "", "",false,false); parameters.push_back(piters);
         CommandParameter psubsample("subsample", "String", "", "", "", "", "",false,false); parameters.push_back(psubsample);
         CommandParameter pcutoff("cutoff", "Number", "", "10", "", "", "",false,false); parameters.push_back(pcutoff);
@@ -160,6 +161,14 @@ TreeGroupCommand::TreeGroupCommand(string option)  {
                                        //if the user has not given a path then, add inputdir. else leave path alone.
                                        if (path == "") {       parameters["name"] = inputDir + it->second;             }
                                }
+                
+                it = parameters.find("count");
+                               //user has given a template file
+                               if(it != parameters.end()){ 
+                                       path = m->hasPath(it->second);
+                                       //if the user has not given a path then, add inputdir. else leave path alone.
+                                       if (path == "") {       parameters["count"] = inputDir + it->second;            }
+                               }
                        }
                        
                        //check for required parameters
@@ -182,6 +191,11 @@ TreeGroupCommand::TreeGroupCommand(string option)  {
                        if (namefile == "not open") { abort = true; }   
                        else if (namefile == "not found") { namefile = ""; }
                        else { m->setNameFile(namefile); }
+            
+            countfile = validParameter.validFile(parameters, "count", true);
+                       if (countfile == "not open") { abort = true; countfile = ""; }  
+                       else if (countfile == "not found") { countfile = ""; }
+                       else { m->setCountTableFile(countfile); }
                        
                        if ((phylipfile == "") && (columnfile == "") && (sharedfile == "")) { 
                                //is there are current file available for either of these?
@@ -204,15 +218,20 @@ TreeGroupCommand::TreeGroupCommand(string option)  {
                        else if ((phylipfile != "") && (columnfile != "")) { m->mothurOut("When running the tree.shared command with a distance file you may not use both the column and the phylip parameters."); m->mothurOutEndLine(); abort = true; }
                        
                        if (columnfile != "") {
-                               if (namefile == "") 
+                               if ((namefile == "") && (countfile == ""))
                                        namefile = m->getNameFile(); 
                                        if (namefile != "") {  m->mothurOut("Using " + namefile + " as input file for the name parameter."); m->mothurOutEndLine(); }
                                        else { 
-                                               m->mothurOut("You need to provide a namefile if you are going to use the column format."); m->mothurOutEndLine(); 
-                                               abort = true; 
+                                               countfile = m->getCountTableFile();
+                        if (countfile != "") {  m->mothurOut("Using " + countfile + " as input file for the count parameter."); m->mothurOutEndLine(); }
+                        else { 
+                            m->mothurOut("You need to provide a namefile or countfile if you are going to use the column format."); m->mothurOutEndLine(); 
+                            abort = true; 
+                        }      
                                        }       
                                }
                        }
+
                        
                        //check for optional parameter and set defaults
                        // ...at some point should added some additional type checking...
@@ -286,8 +305,8 @@ TreeGroupCommand::TreeGroupCommand(string option)  {
 TreeGroupCommand::~TreeGroupCommand(){
        if (abort == false) {
                if (format == "sharedfile") {  delete input; }
-               else { delete readMatrix;  delete matrix; delete list; }
-               delete tmap;  
+               else { delete list; }
+               delete ct;  
        }
        
 }
@@ -400,8 +419,16 @@ int TreeGroupCommand::execute(){
                        m->runParse = false;
                        
                        //create treemap class from groupmap for tree class to use
-                       tmap = new TreeMap();
-                       tmap->makeSim(m->getAllGroups());
+                       ct = new CountTable();
+            set<string> nameMap;
+            map<string, string> groupMap;
+            set<string> gps;
+            for (int i = 0; i < m->getAllGroups().size(); i++) { 
+                nameMap.insert(m->getAllGroups()[i]); 
+                gps.insert(m->getAllGroups()[i]); 
+                groupMap[m->getAllGroups()[i]] = m->getAllGroups()[i];
+            }
+            ct->createTable(nameMap, groupMap, gps);
                        
                        //clear globaldatas old tree names if any
                        m->Treenames.clear();
@@ -418,37 +445,47 @@ int TreeGroupCommand::execute(){
                }else{
                        //read in dist file
                        filename = inputfile;
-               
+            
+            ReadMatrix* readMatrix;
                        if (format == "column") { readMatrix = new ReadColumnMatrix(filename); }        
                        else if (format == "phylip") { readMatrix = new ReadPhylipMatrix(filename); }
                                
                        readMatrix->setCutoff(cutoff);
        
-                       if(namefile != ""){     
-                               nameMap = new NameAssignment(namefile);
-                               nameMap->readMap();
-                       }
-                       else{
-                               nameMap = NULL;
-                       }
-       
-                       readMatrix->read(nameMap);
+            ct = NULL;
+            if(namefile != ""){        
+                nameMap = new NameAssignment(namefile);
+                nameMap->readMap();
+                readMatrix->read(nameMap);
+            }else if (countfile != "") {
+                ct = new CountTable();
+                ct->readTable(countfile);
+                readMatrix->read(ct);
+            }
+
                        list = readMatrix->getListVector();
-                       matrix = readMatrix->getMatrix();
+                       SparseDistanceMatrix* dMatrix = readMatrix->getDMatrix();
 
                        //make treemap
-                       tmap = new TreeMap();
-                       
-                       if (m->control_pressed) { return 0; }
-                       
-                       tmap->makeSim(list);
+            if (ct != NULL) { delete ct; }
+                       ct = new CountTable();
+            set<string> nameMap;
+            map<string, string> groupMap;
+            set<string> gps;
+            for (int i = 0; i < list->getNumBins(); i++) {
+                string bin = list->get(i);
+                nameMap.insert(bin); 
+                gps.insert(bin); 
+                groupMap[bin] = bin;
+            }
+            ct->createTable(nameMap, groupMap, gps);
                        
-                       vector<string> namesGroups = tmap->getNamesOfGroups();
+                       vector<string> namesGroups = ct->getNamesOfGroups();
                        m->setGroups(namesGroups);
                
                        //clear globaldatas old tree names if any
                        m->Treenames.clear();
-               
+            
                        //fills globaldatas tree names
                        m->Treenames = m->getGroups();
                        
@@ -457,7 +494,9 @@ int TreeGroupCommand::execute(){
                        
                        if (m->control_pressed) { return 0; }
                        
-                       vector< vector<double> > matrix = makeSimsDist();
+                       vector< vector<double> > matrix = makeSimsDist(dMatrix);
+            delete readMatrix;
+            delete dMatrix;
                        
                        if (m->control_pressed) { return 0; }
 
@@ -502,13 +541,12 @@ int TreeGroupCommand::execute(){
 Tree* TreeGroupCommand::createTree(vector< vector<double> >& simMatrix){
        try {
                //create tree
-               t = new Tree(tmap, simMatrix);
+               t = new Tree(ct, simMatrix);
         
         if (m->control_pressed) { delete t; t = NULL; return t; }
                
         //assemble tree
-        map<string, string> empty;
-               t->assembleTree(empty);
+               t->assembleTree();
 
                return t;
        }
@@ -554,7 +592,7 @@ void TreeGroupCommand::printSims(ostream& out, vector< vector<double> >& simMatr
        }
 }
 /***********************************************************/
-vector< vector<double> > TreeGroupCommand::makeSimsDist() {
+vector< vector<double> > TreeGroupCommand::makeSimsDist(SparseDistanceMatrix* matrix) {
        try {
                numGroups = list->size();
                
@@ -569,13 +607,17 @@ vector< vector<double> > TreeGroupCommand::makeSimsDist() {
                
                //go through sparse matrix and fill sims
                //go through each cell in the sparsematrix
-               for(MatData currentCell = matrix->begin(); currentCell != matrix->end(); currentCell++){
-                       //similairity = -(distance-1)
-                       simMatrix[currentCell->row][currentCell->column] = -(currentCell->dist -1.0);   
-                       simMatrix[currentCell->column][currentCell->row] = -(currentCell->dist -1.0);   
-                       
-                       if (m->control_pressed) { return simMatrix; }
+        for (int i = 0; i < matrix->seqVec.size(); i++) {
+            for (int j = 0; j < matrix->seqVec[i].size(); j++) {
+                
+                //already checked everyone else in row
+                if (i < matrix->seqVec[i][j].index) {   
+                    simMatrix[i][matrix->seqVec[i][j].index] = -(matrix->seqVec[i][j].dist -1.0);      
+                    simMatrix[matrix->seqVec[i][j].index][i] = -(matrix->seqVec[i][j].dist -1.0);      
                        
+                    if (m->control_pressed) { return simMatrix; }
+                }
+            }
                }
 
                return simMatrix;