]> git.donarmstrong.com Git - mothur.git/blobdiff - tree.cpp
added get.groups and remove.groups commands
[mothur.git] / tree.cpp
index 400a72ae388a88c5b94aa2b8a717471f600b33ed..f8bb76fd664383f148c73eca62aa5cc2291202c8 100644 (file)
--- a/tree.cpp
+++ b/tree.cpp
@@ -9,7 +9,19 @@
 
 #include "tree.h"
 
-
+/*****************************************************************/
+Tree::Tree(string g) {
+       try {
+               globaldata = GlobalData::getInstance();
+               m = MothurOut::getInstance();
+               
+               parseTreeFile();  globaldata->runParse = false;  
+       }
+       catch(exception& e) {
+               m->errorOut(e, "Tree", "Tree - just parse");
+               exit(1);
+       }
+}
 /*****************************************************************/
 Tree::Tree() {
        try {
@@ -22,17 +34,28 @@ Tree::Tree() {
                numNodes = 2*numLeaves - 1;
                
                tree.resize(numNodes);
+               
+               //initialize groupNodeInfo
+               for (int i = 0; i < globaldata->gTreemap->namesOfGroups.size(); i++) {
+                       groupNodeInfo[globaldata->gTreemap->namesOfGroups[i]].resize(0);
+               }
 
                //initialize tree with correct number of nodes, name and group info.
                for (int i = 0; i < numNodes; i++) {
                        //initialize leaf nodes
                        if (i <= (numLeaves-1)) {
                                tree[i].setName(globaldata->Treenames[i]);
-                               vector<string> tempGroups; tempGroups.push_back(globaldata->gTreemap->getGroup(globaldata->Treenames[i]));
+                               
+                               //save group info
+                               string group = globaldata->gTreemap->getGroup(globaldata->Treenames[i]);
+                               vector<string> tempGroups; tempGroups.push_back(group);
                                tree[i].setGroup(tempGroups);
+                               groupNodeInfo[group].push_back(i); 
+                               
                                //set pcount and pGroup for groupname to 1.
-                               tree[i].pcount[globaldata->gTreemap->getGroup(globaldata->Treenames[i])] = 1;
-                               tree[i].pGroups[globaldata->gTreemap->getGroup(globaldata->Treenames[i])] = 1;
+                               tree[i].pcount[group] = 1;
+                               tree[i].pGroups[group] = 1;
+                               
                                //Treemap knows name, group and index to speed up search
                                globaldata->gTreemap->setIndex(globaldata->Treenames[i], i);
        
@@ -69,6 +92,9 @@ void Tree::addNamesToCounts() {
 
                                
                //go through each leaf and update its pcounts and pgroups
+               
+               //float A = clock();
+
                for (int i = 0; i < numLeaves; i++) {
 
                        string name = tree[i].getName();
@@ -78,14 +104,18 @@ void Tree::addNamesToCounts() {
                        if (itNames == globaldata->names.end()) { m->mothurOut(name + " is not in your name file, please correct."); m->mothurOutEndLine(); exit(1);  }
                        else {
                                vector<string> dupNames;
-                               splitAtComma(globaldata->names[name], dupNames);
+                               m->splitAtComma(globaldata->names[name], dupNames);
                                
                                map<string, int>::iterator itCounts;
                                int maxPars = 1;
+                               set<string> groupsAddedForThisNode;
                                for (int j = 0; j < dupNames.size(); j++) {
-                               
+                                       
+                                       string group = globaldata->gTreemap->getGroup(dupNames[j]);
+                                       
                                        if (dupNames[j] != name) {//you already added yourself in the constructor
-                                               string group = globaldata->gTreemap->getGroup(dupNames[j]);
+                               
+                                               if (groupsAddedForThisNode.count(group) == 0)  {  groupNodeInfo[group].push_back(i);  groupsAddedForThisNode.insert(group);  } //if you have not already added this node for this group, then add it
                                                
                                                //update pcounts
                                                itCounts = tree[i].pcount.find(group);
@@ -107,7 +137,7 @@ void Tree::addNamesToCounts() {
                                                if(tree[i].pGroups[group] > maxPars){
                                                        maxPars = tree[i].pGroups[group];
                                                }
-                                       }//end if
+                                       }else {  groupsAddedForThisNode.insert(group);  } //add it so you don't add it to groupNodeInfo again
                                }//end for
                                
                                if (maxPars > 1) { //then we have some more dominant groups
@@ -132,7 +162,11 @@ void Tree::addNamesToCounts() {
                                tree[i].setGroup(nodeGroups);
                                
                        }//end else
-               }//end for                                      
+               }//end for              
+               
+               //float B = clock();
+               //cout << "addNamesToCounts\t" << (B - A) / CLOCKS_PER_SEC << endl;     
+
        }
        catch(exception& e) {
                m->errorOut(e, "Tree", "addNamesToCounts");
@@ -168,7 +202,8 @@ void Tree::setIndex(string searchName, int index) {
 /*****************************************************************/
 int Tree::assembleTree() {
        try {
-       
+               //float A = clock();
+
                //if user has given a names file we want to include that info in the pgroups and pcount info.
                if(globaldata->names.size() != 0) {  addNamesToCounts();  }
                
@@ -179,7 +214,28 @@ int Tree::assembleTree() {
                        tree[i].pGroups = (mergeGroups(i));
                        tree[i].pcount = (mergeGcounts(i));
                }
+               //float B = clock();
+               //cout << "assembleTree\t" << (B-A) / CLOCKS_PER_SEC << endl;
+               return 0;
+       }
+       catch(exception& e) {
+               m->errorOut(e, "Tree", "assembleTree");
+               exit(1);
+       }
+}
+/*****************************************************************/
+int Tree::assembleTree(string n) {
+       try {
                
+               //build the pGroups in non leaf nodes to be used in the parsimony calcs.
+               for (int i = numLeaves; i < numNodes; i++) {
+                       if (m->control_pressed) { return 1; }
+
+                       tree[i].pGroups = (mergeGroups(i));
+                       tree[i].pcount = (mergeGcounts(i));
+               }
+               //float B = clock();
+               //cout << "assembleTree\t" << (B-A) / CLOCKS_PER_SEC << endl;
                return 0;
        }
        catch(exception& e) {
@@ -187,6 +243,7 @@ int Tree::assembleTree() {
                exit(1);
        }
 }
+
 /*****************************************************************/
 void Tree::getCopy(Tree* copy) {
        try {
@@ -219,6 +276,8 @@ void Tree::getCopy(Tree* copy) {
                        tree[i].pcount = copy->tree[i].pcount;
                }
                
+               groupNodeInfo = copy->groupNodeInfo;
+               
        }
        catch(exception& e) {
                m->errorOut(e, "Tree", "getCopy");
@@ -292,14 +351,14 @@ map<string, int> Tree::mergeUserGroups(int i, vector<string> g) {
                
                //loop through nodes groups removing the ones the user doesn't want
                for(it=tree[lc].pGroups.begin();it!=tree[lc].pGroups.end();){
-                               if (inUsersGroups(it->first, g) != true) {
+                               if (m->inUsersGroups(it->first, g) != true) {
                                        tree[lc].pGroups.erase(it++);
                                }else { it++; }
                }
 
                //loop through nodes groups removing the ones the user doesn't want
                for(it=tree[rc].pGroups.begin();it!=tree[rc].pGroups.end();){
-                               if (inUsersGroups(it->first, g) != true) {
+                               if (m->inUsersGroups(it->first, g) != true) {
                                        tree[rc].pGroups.erase(it++);
                                }else { it++; }
                }
@@ -371,6 +430,11 @@ map<string,int> Tree::mergeGcounts(int position) {
 
 void Tree::randomLabels(vector<string> g) {
        try {
+       
+               //initialize groupNodeInfo
+               for (int i = 0; i < globaldata->gTreemap->namesOfGroups.size(); i++) {
+                       groupNodeInfo[globaldata->gTreemap->namesOfGroups[i]].resize(0);
+               }
                
                for(int i = 0; i < numLeaves; i++){
                        int z;
@@ -381,8 +445,8 @@ void Tree::randomLabels(vector<string> g) {
                        //if either of the leaf nodes you are about to switch are not in the users groups then you don't want to switch them.
                        bool treez, treei;
                
-                       treez = inUsersGroups(tree[z].getGroup(), g);
-                       treei = inUsersGroups(tree[i].getGroup(), g);
+                       treez = m->inUsersGroups(tree[z].getGroup(), g);
+                       treei = m->inUsersGroups(tree[i].getGroup(), g);
                        
                        if ((treez == true) && (treei == true)) {
                                //switches node i and node z's info.
@@ -402,6 +466,9 @@ void Tree::randomLabels(vector<string> g) {
                                tree[z].pcount = (tree[i].pcount);
                                tree[i].pcount = (gcount_hold);
                        }
+                       
+                       for (int k = 0; k < (tree[i].getGroup()).size(); k++) {  groupNodeInfo[(tree[i].getGroup())[k]].push_back(i); }
+                       for (int k = 0; k < (tree[z].getGroup()).size(); k++) {  groupNodeInfo[(tree[z].getGroup())[k]].push_back(z); }
                }
        }
        catch(exception& e) {
@@ -458,14 +525,14 @@ void Tree::randomBlengths()  {
 /*************************************************************************************************/
 void Tree::assembleRandomUnifracTree(vector<string> g) {
        randomLabels(g);
-       assembleTree();
+       assembleTree("noNameCounts");
 }
 /*************************************************************************************************/
 void Tree::assembleRandomUnifracTree(string groupA, string groupB) {
 
        vector<string> temp; temp.push_back(groupA); temp.push_back(groupB);
        randomLabels(temp);
-       assembleTree();
+       assembleTree("noNameCounts");
 }
 
 /*************************************************************************************************/
@@ -542,10 +609,10 @@ void Tree::printForBoot(ostream& out) {
 void Tree::createNewickFile(string f) {
        try {
                int root = findRoot();
-               //filename = getRootName(globaldata->getTreeFile()) + "newick";
+               //filename = m->getRootName(globaldata->getTreeFile()) + "newick";
                filename = f;
 
-               openOutputFile(filename, out);
+               m->openOutputFile(filename, out);
                
                printBranch(root, out, "branch");
                
@@ -643,7 +710,7 @@ void Tree::parseTreeFile() {
        try {
                string filename = globaldata->getTreeFile();
                ifstream filehandle;
-               openInputFile(filename, filehandle);
+               m->openInputFile(filename, filehandle);
                int c, comment;
                comment = 0;
                int done = 1;
@@ -736,20 +803,39 @@ int Tree::readTreeString(ifstream& filehandle)    {
 //cout << " at beginning of while " <<  k << endl;                     
                        if(c == ')')  {    
                                //to pass over labels in trees
-                               string label = readLabel(filehandle);
+                               c=filehandle.get();
+                               while((c!=',') && (c != -1) && (c!= ':') && (c!=';')){ c=filehandle.get(); }
+                               filehandle.putback(c);
                        }
-                       
                        if(c == ';') { return 0; }
                        if(c == -1) { return 0; }
-                       
                        //if you are a name
                        if((c != '(') && (c != ')') && (c != ',') && (c != ':') && (c != '\n') && (c != '\t') && (c != 32)) { //32 is space
-                               name = readName(filehandle);
+                               name = "";
+                               c = filehandle.get();
+                       //k = c;
+//cout << k << endl;
+                               while ((c != '(') && (c != ')') && (c != ',') && (c != ':') && (c != '\n') && (c != 32) && (c != '\t')) {                       
+                                       name += c;
+                                       c = filehandle.get();
+                       //k = c;
+//cout << " in name while " << k << endl;
+                               }
+                               
+//cout << "name = " << name << endl;
                                globaldata->Treenames.push_back(name);
+                               filehandle.putback(c);
+//k = c;
+//cout << " after putback" <<  k << endl;
                        } 
                        
                        if(c  == ':') { //read until you reach the end of the branch length
-                               string bl = readBranchLength(filehandle);
+                               while ((c != '(') && (c != ')') && (c != ',') && (c != ';') && (c != '\n') && (c != '\t') && (c != 32)) {
+                                       c = filehandle.get();
+       //k = c;
+       //cout << " in branch while " << k << endl;
+                               }
+                               filehandle.putback(c);
                        }
                
                        c = filehandle.get();
@@ -770,68 +856,6 @@ int Tree::readTreeString(ifstream& filehandle)     {
 }      
 
 /*******************************************************/
-string Tree::readLabel(ifstream& filehandle)   {
-       try {
-               
-               string label = "";
-               
-               //to pass over labels in trees
-               int c=filehandle.get();
-               while((c!=',') && (c != -1) && (c!= ':') && (c!=';')){ label += c; c=filehandle.get(); }
-               filehandle.putback(c);
-               
-               return label;
-               
-       }
-       catch(exception& e) {
-               m->errorOut(e, "Tree", "readLabel");
-               exit(1);
-       }
-}      
-/*******************************************************/
-string Tree::readName(ifstream& filehandle)    {
-       try {
-               
-               string name = "";
-               int c = filehandle.get();
-               
-               while ((c != '(') && (c != ')') && (c != ',') && (c != ':') && (c != '\n') && (c != 32) && (c != '\t')) {                       
-                       name += c;
-                       c = filehandle.get();
-               }
-                               
-//cout << "name = " << name << endl;
-               filehandle.putback(c);
-               
-               return name;
-               
-       }
-       catch(exception& e) {
-               m->errorOut(e, "Tree", "readName");
-               exit(1);
-       }
-}      
-/*******************************************************/
-string Tree::readBranchLength(ifstream& filehandle)    {
-       try {
-               
-               string br = "";
-               int c;
-               while ((c != '(') && (c != ')') && (c != ',') && (c != ';') && (c != '\n') && (c != '\t') && (c != 32)) {
-                       br += c;
-                       c = filehandle.get();
-               }
-               filehandle.putback(c);
-               
-               return br;
-               
-       }
-       catch(exception& e) {
-               m->errorOut(e, "Tree", "readBranchLength");
-               exit(1);
-       }
-}      
 
 /*******************************************************/
-/*******************************************************/