X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=tree.cpp;h=d9d71aea2dd1859d6e198302325fb51d2fdf9367;hb=d6c0a11d1cecfac18b323285e7ffadb7f58e848f;hp=06f92cf46e22cbb28f4ff28596977d5b62fb7e02;hpb=edad2fa8defaa6856b3606a215bf64b91340eeb4;p=mothur.git diff --git a/tree.cpp b/tree.cpp index 06f92cf..d9d71ae 100644 --- a/tree.cpp +++ b/tree.cpp @@ -10,9 +10,8 @@ #include "tree.h" /*****************************************************************/ -Tree::Tree(int num) { +Tree::Tree(int num, TreeMap* t) : tmap(t) { try { - globaldata = GlobalData::getInstance(); m = MothurOut::getInstance(); numLeaves = num; @@ -26,12 +25,13 @@ Tree::Tree(int num) { } } /*****************************************************************/ -Tree::Tree(string g) { +Tree::Tree(string g) { //do not use tree generated by this its just to extract the treenames, its a chicken before the egg thing that needs to be revisited. try { - globaldata = GlobalData::getInstance(); m = MothurOut::getInstance(); - parseTreeFile(); globaldata->runParse = false; + tmap = NULL; + + parseTreeFile(); m->runParse = false; } catch(exception& e) { m->errorOut(e, "Tree", "Tree - just parse"); @@ -39,31 +39,30 @@ Tree::Tree(string g) { } } /*****************************************************************/ -Tree::Tree() { +Tree::Tree(TreeMap* t) : tmap(t) { try { - globaldata = GlobalData::getInstance(); m = MothurOut::getInstance(); - if (globaldata->runParse == true) { parseTreeFile(); globaldata->runParse = false; } + if (m->runParse == true) { parseTreeFile(); m->runParse = false; } //for(int i = 0; i < globaldata->Treenames.size(); i++) { cout << i << '\t' << globaldata->Treenames[i] << endl; } - numLeaves = globaldata->Treenames.size(); + numLeaves = m->Treenames.size(); 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); + for (int i = 0; i < (tmap->getNamesOfGroups()).size(); i++) { + groupNodeInfo[(tmap->getNamesOfGroups())[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]); + tree[i].setName(m->Treenames[i]); //save group info - string group = globaldata->gTreemap->getGroup(globaldata->Treenames[i]); + string group = tmap->getGroup(m->Treenames[i]); vector tempGroups; tempGroups.push_back(group); tree[i].setGroup(tempGroups); @@ -74,7 +73,7 @@ Tree::Tree() { tree[i].pGroups[group] = 1; //Treemap knows name, group and index to speed up search - globaldata->gTreemap->setIndex(globaldata->Treenames[i], i); + tmap->setIndex(m->Treenames[i], i); //intialize non leaf nodes }else if (i > (numLeaves-1)) { @@ -117,19 +116,19 @@ void Tree::addNamesToCounts() { string name = tree[i].getName(); - map::iterator itNames = globaldata->names.find(name); + map::iterator itNames = m->names.find(name); - if (itNames == globaldata->names.end()) { m->mothurOut(name + " is not in your name file, please correct."); m->mothurOutEndLine(); exit(1); } + if (itNames == m->names.end()) { m->mothurOut(name + " is not in your name file, please correct."); m->mothurOutEndLine(); exit(1); } else { vector dupNames; - m->splitAtComma(globaldata->names[name], dupNames); + m->splitAtComma(m->names[name], dupNames); map::iterator itCounts; int maxPars = 1; set groupsAddedForThisNode; for (int j = 0; j < dupNames.size(); j++) { - string group = globaldata->gTreemap->getGroup(dupNames[j]); + string group = tmap->getGroup(dupNames[j]); if (dupNames[j] != name) {//you already added yourself in the constructor @@ -196,7 +195,7 @@ int Tree::getIndex(string searchName) { try { //Treemap knows name, group and index to speed up search // getIndex function will return the vector index or -1 if seq is not found. - int index = globaldata->gTreemap->getIndex(searchName); + int index = tmap->getIndex(searchName); return index; } @@ -210,7 +209,7 @@ int Tree::getIndex(string searchName) { void Tree::setIndex(string searchName, int index) { try { //set index in treemap - globaldata->gTreemap->setIndex(searchName, index); + tmap->setIndex(searchName, index); } catch(exception& e) { m->errorOut(e, "Tree", "setIndex"); @@ -223,7 +222,7 @@ int Tree::assembleTree() { //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(); } + if(m->names.size() != 0) { addNamesToCounts(); } //build the pGroups in non leaf nodes to be used in the parsimony calcs. for (int i = numLeaves; i < numNodes; i++) { @@ -274,7 +273,7 @@ void Tree::getSubTree(Tree* copy, vector Groups) { tree[i].setName(Groups[i]); //save group info - string group = globaldata->gTreemap->getGroup(Groups[i]); + string group = tmap->getGroup(Groups[i]); vector tempGroups; tempGroups.push_back(group); tree[i].setGroup(tempGroups); groupNodeInfo[group].push_back(i); @@ -284,7 +283,7 @@ void Tree::getSubTree(Tree* copy, vector Groups) { tree[i].pGroups[group] = 1; //Treemap knows name, group and index to speed up search - globaldata->gTreemap->setIndex(Groups[i], i); + tmap->setIndex(Groups[i], i); //intialize non leaf nodes }else if (i > (numLeaves-1)) { @@ -330,7 +329,9 @@ void Tree::getSubTree(Tree* copy, vector Groups) { copy->tree[i].setParent(grandparent); copy->tree[i].setBranchLength((copy->tree[i].getBranchLength()+copy->tree[parent].getBranchLength())); - copy->tree[grandparent].setChildren(grandparentLC, grandparentRC); + if (grandparent != -1) { + copy->tree[grandparent].setChildren(grandparentLC, grandparentRC); + } removedLeaves.insert(sibIndex); } }else{ @@ -357,7 +358,9 @@ void Tree::getSubTree(Tree* copy, vector Groups) { copy->tree[sibIndex].setParent(grandparent); copy->tree[sibIndex].setBranchLength((copy->tree[sibIndex].getBranchLength()+copy->tree[parent].getBranchLength())); - copy->tree[grandparent].setChildren(grandparentLC, grandparentRC); + if (grandparent != -1) { + copy->tree[grandparent].setChildren(grandparentLC, grandparentRC); + } removedLeaves.insert(i); }else{ //neither of us are, so we want to eliminate ourselves and our parent @@ -367,8 +370,11 @@ void Tree::getSubTree(Tree* copy, vector Groups) { int parentsSibIndex; if (grandparent != -1) { int greatgrandparent = copy->tree[grandparent].getParent(); - int greatgrandparentLC = copy->tree[greatgrandparent].getLChild(); - int greatgrandparentRC = copy->tree[greatgrandparent].getRChild(); + int greatgrandparentLC, greatgrandparentRC; + if (greatgrandparent != -1) { + greatgrandparentLC = copy->tree[greatgrandparent].getLChild(); + greatgrandparentRC = copy->tree[greatgrandparent].getRChild(); + } int grandparentLC = copy->tree[grandparent].getLChild(); int grandparentRC = copy->tree[grandparent].getRChild(); @@ -382,10 +388,12 @@ void Tree::getSubTree(Tree* copy, vector Groups) { copy->tree[parentsSibIndex].setParent(greatgrandparent); copy->tree[parentsSibIndex].setBranchLength((copy->tree[parentsSibIndex].getBranchLength()+copy->tree[grandparent].getBranchLength())); - copy->tree[greatgrandparent].setChildren(greatgrandparentLC, greatgrandparentRC); + if (greatgrandparent != -1) { + copy->tree[greatgrandparent].setChildren(greatgrandparentLC, greatgrandparentRC); + } }else{ - copy->tree[parent].setChildren(-1, -1); - cout << "issues with making subtree" << endl; + copy->tree[parent].setParent(-1); + //cout << "issues with making subtree" << endl; } removedLeaves.insert(sibIndex); removedLeaves.insert(i); @@ -403,7 +411,6 @@ void Tree::getSubTree(Tree* copy, vector Groups) { int nextSpot = numLeaves; populateNewTree(copy->tree, root, nextSpot); - } catch(exception& e) { m->errorOut(e, "Tree", "getCopy"); @@ -428,7 +435,7 @@ int Tree::populateNewTree(vector& oldtree, int node, int& index) { return (index++); }else { //you are a leaf - int indexInNewTree = globaldata->gTreemap->getIndex(oldtree[node].getName()); + int indexInNewTree = tmap->getIndex(oldtree[node].getName()); return indexInNewTree; } } @@ -625,8 +632,8 @@ void Tree::randomLabels(vector 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 < (tmap->getNamesOfGroups()).size(); i++) { + groupNodeInfo[(tmap->getNamesOfGroups())[i]].resize(0); } for(int i = 0; i < numLeaves; i++){ @@ -699,7 +706,7 @@ void Tree::randomLabels(string groupA, string groupB) { exit(1); } } -/**************************************************************************************************/ +**************************************************************************************************/ void Tree::randomBlengths() { try { for(int i=numNodes-1;i>=0;i--){ @@ -801,7 +808,7 @@ void Tree::print(ostream& out, string mode) { void Tree::createNewickFile(string f) { try { int root = findRoot(); - //filename = m->getRootName(globaldata->getTreeFile()) + "newick"; + filename = f; m->openOutputFile(filename, out); @@ -865,9 +872,14 @@ try { if (tree[node].getBranchLength() != -1) { out << ":" << tree[node].getBranchLength(); } + }else if (mode == "deunique") { + //if there is a branch length then print it + if (tree[node].getBranchLength() != -1) { + out << ":" << tree[node].getBranchLength(); + } } }else { //you are a leaf - string leafGroup = globaldata->gTreemap->getGroup(tree[node].getName()); + string leafGroup = tmap->getGroup(tree[node].getName()); if (mode == "branch") { out << leafGroup; @@ -890,6 +902,53 @@ try { if (tree[node].getBranchLength() != -1) { out << ":" << tree[node].getBranchLength(); } + }else if (mode == "deunique") { + map::iterator itNames = m->names.find(tree[node].getName()); + + string outputString = ""; + if (itNames != m->names.end()) { + + vector dupNames; + m->splitAtComma((itNames->second), dupNames); + + if (dupNames.size() == 1) { + outputString += tree[node].getName(); + if (tree[node].getBranchLength() != -1) { + outputString += ":" + toString(tree[node].getBranchLength()); + } + }else { + outputString += "("; + + for (int u = 0; u < dupNames.size()-1; u++) { + outputString += dupNames[u]; + + if (tree[node].getBranchLength() != -1) { + outputString += ":" + toString(0.0); + } + outputString += ","; + } + + outputString += dupNames[dupNames.size()-1]; + if (tree[node].getBranchLength() != -1) { + outputString += ":" + toString(0.0); + } + + outputString += ")"; + if (tree[node].getBranchLength() != -1) { + outputString += ":" + toString(tree[node].getBranchLength()); + } + } + }else { + outputString = tree[node].getName(); + //if there is a branch length then print it + if (tree[node].getBranchLength() != -1) { + outputString += ":" + toString(tree[node].getBranchLength()); + } + + m->mothurOut("[ERROR]: " + tree[node].getName() + " is not in your namefile, please correct."); m->mothurOutEndLine(); + } + + out << outputString; } } @@ -930,7 +989,7 @@ void Tree::printBranch(int node, ostream& out, string mode, vector& theseN } } }else { //you are a leaf - string leafGroup = globaldata->gTreemap->getGroup(theseNodes[node].getName()); + string leafGroup = tmap->getGroup(theseNodes[node].getName()); if (mode == "branch") { out << leafGroup; @@ -979,7 +1038,7 @@ void Tree::parseTreeFile() { //only takes names from the first tree and assumes that all trees use the same names. try { - string filename = globaldata->getTreeFile(); + string filename = m->getTreeFile(); ifstream filehandle; m->openInputFile(filename, filehandle); int c, comment; @@ -1045,9 +1104,9 @@ void Tree::parseTreeFile() { //c = , until done with translation then c = ; h = name.substr(name.length()-1, name.length()); name.erase(name.end()-1); //erase the comma - globaldata->Treenames.push_back(number); + m->Treenames.push_back(number); } - if(number == ";") { globaldata->Treenames.pop_back(); } //in case ';' from translation is on next line instead of next to last name + if(number == ";") { m->Treenames.pop_back(); } //in case ';' from translation is on next line instead of next to last name } } filehandle.close(); @@ -1094,7 +1153,7 @@ int Tree::readTreeString(ifstream& filehandle) { } //cout << "name = " << name << endl; - globaldata->Treenames.push_back(name); + m->Treenames.push_back(name); filehandle.putback(c); //k = c; //cout << " after putback" << k << endl;