X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=tree.cpp;h=eca53d5527bbdda6dbc97ddc0c9f4e7766e15d25;hb=74844a60d80c6dd06e3fb02ee9b928424f9019b0;hp=1e115bddad1a451c181c4c26b90b758788df1155;hpb=dbc9352256135d7b3e2b17dddc5c506770e2df92;p=mothur.git diff --git a/tree.cpp b/tree.cpp index 1e115bd..eca53d5 100644 --- a/tree.cpp +++ b/tree.cpp @@ -13,9 +13,12 @@ /*****************************************************************/ Tree::Tree() { try { - globaldata = GlobalData::getInstance(); - numLeaves = globaldata->gTreemap->getNumSeqs(); + m = MothurOut::getInstance(); + + if (globaldata->runParse == true) { parseTreeFile(); globaldata->runParse = false; } +//for(int i = 0; i < globaldata->Treenames.size(); i++) { cout << i << '\t' << globaldata->Treenames[i] << endl; } + numLeaves = globaldata->Treenames.size(); numNodes = 2*numLeaves - 1; tree.resize(numNodes); @@ -24,31 +27,118 @@ Tree::Tree() { for (int i = 0; i < numNodes; i++) { //initialize leaf nodes if (i <= (numLeaves-1)) { - tree[i].setName(globaldata->gTreemap->namesOfSeqs[i]); - tree[i].setGroup(globaldata->gTreemap->getGroup(globaldata->gTreemap->namesOfSeqs[i])); + tree[i].setName(globaldata->Treenames[i]); + vector tempGroups; tempGroups.push_back(globaldata->gTreemap->getGroup(globaldata->Treenames[i])); + tree[i].setGroup(tempGroups); //set pcount and pGroup for groupname to 1. - tree[i].pcount[globaldata->gTreemap->getGroup(globaldata->gTreemap->namesOfSeqs[i])] = 1; - tree[i].pGroups[globaldata->gTreemap->getGroup(globaldata->gTreemap->namesOfSeqs[i])] = 1; + tree[i].pcount[globaldata->gTreemap->getGroup(globaldata->Treenames[i])] = 1; + tree[i].pGroups[globaldata->gTreemap->getGroup(globaldata->Treenames[i])] = 1; //Treemap knows name, group and index to speed up search - globaldata->gTreemap->setIndex(globaldata->gTreemap->namesOfSeqs[i], i); + globaldata->gTreemap->setIndex(globaldata->Treenames[i], i); //intialize non leaf nodes }else if (i > (numLeaves-1)) { tree[i].setName(""); - tree[i].setGroup(""); + vector tempGroups; + tree[i].setGroup(tempGroups); } } } catch(exception& e) { - cout << "Standard Error: " << e.what() << " has occurred in the Tree class Function Tree. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + m->errorOut(e, "Tree", "Tree"); exit(1); } - catch(...) { - cout << "An unknown error has occurred in the Tree class function Tree. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; - exit(1); - } } +/*****************************************************************/ +Tree::~Tree() {} +/*****************************************************************/ +void Tree::addNamesToCounts() { + try { + //ex. seq1 seq2,seq3,se4 + // seq1 = pasture + // seq2 = forest + // seq4 = pasture + // seq3 = ocean + + //before this function seq1.pcount = pasture -> 1 + //after seq1.pcount = pasture -> 2, forest -> 1, ocean -> 1 + + //before this function seq1.pgroups = pasture -> 1 + //after seq1.pgroups = pasture -> 1 since that is the dominant group + + + //go through each leaf and update its pcounts and pgroups + for (int i = 0; i < numLeaves; i++) { + string name = tree[i].getName(); + + map::iterator itNames = globaldata->names.find(name); + + if (itNames == globaldata->names.end()) { m->mothurOut(name + " is not in your name file, please correct."); m->mothurOutEndLine(); exit(1); } + else { + vector dupNames; + splitAtComma(globaldata->names[name], dupNames); + + map::iterator itCounts; + int maxPars = 1; + for (int j = 0; j < dupNames.size(); j++) { + + if (dupNames[j] != name) {//you already added yourself in the constructor + string group = globaldata->gTreemap->getGroup(dupNames[j]); + + //update pcounts + itCounts = tree[i].pcount.find(group); + if (itCounts == tree[i].pcount.end()) { //new group, add it + tree[i].pcount[group] = 1; + }else { + tree[i].pcount[group]++; + } + + //update pgroups + itCounts = tree[i].pGroups.find(group); + if (itCounts == tree[i].pGroups.end()) { //new group, add it + tree[i].pGroups[group] = 1; + }else { + tree[i].pGroups[group]++; + } + + //keep highest group + if(tree[i].pGroups[group] > maxPars){ + maxPars = tree[i].pGroups[group]; + } + }//end if + }//end for + + if (maxPars > 1) { //then we have some more dominant groups + //erase all the groups that are less than maxPars because you found a more dominant group. + for(it=tree[i].pGroups.begin();it!=tree[i].pGroups.end();){ + if(it->second < maxPars){ + tree[i].pGroups.erase(it++); + }else { it++; } + } + //set one remaining groups to 1 + for(it=tree[i].pGroups.begin();it!=tree[i].pGroups.end();it++){ + tree[i].pGroups[it->first] = 1; + } + }//end if + + //update groups to reflect all the groups this node represents + vector nodeGroups; + map::iterator itGroups; + for (itGroups = tree[i].pcount.begin(); itGroups != tree[i].pcount.end(); itGroups++) { + nodeGroups.push_back(itGroups->first); + } + tree[i].setGroup(nodeGroups); + + }//end else + }//end for + + } + catch(exception& e) { + m->errorOut(e, "Tree", "addNamesToCounts"); + exit(1); + } +} /*****************************************************************/ int Tree::getIndex(string searchName) { try { @@ -59,13 +149,9 @@ int Tree::getIndex(string searchName) { } catch(exception& e) { - cout << "Standard Error: " << e.what() << " has occurred in the Tree class Function getIndex. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + m->errorOut(e, "Tree", "getIndex"); exit(1); } - catch(...) { - cout << "An unknown error has occurred in the Tree class function getIndex. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; - exit(1); - } } /*****************************************************************/ @@ -75,17 +161,17 @@ void Tree::setIndex(string searchName, int index) { globaldata->gTreemap->setIndex(searchName, index); } catch(exception& e) { - cout << "Standard Error: " << e.what() << " has occurred in the Tree class Function setIndex. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + m->errorOut(e, "Tree", "setIndex"); exit(1); } - catch(...) { - cout << "An unknown error has occurred in the Tree class function setIndex. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; - exit(1); - } } /*****************************************************************/ void Tree::assembleTree() { try { + + //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(); } + //build the pGroups in non leaf nodes to be used in the parsimony calcs. for (int i = numLeaves; i < numNodes; i++) { tree[i].pGroups = (mergeGroups(i)); @@ -93,13 +179,9 @@ void Tree::assembleTree() { } } catch(exception& e) { - cout << "Standard Error: " << e.what() << " has occurred in the Tree class Function assembleTree. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + m->errorOut(e, "Tree", "assembleTree"); exit(1); } - catch(...) { - cout << "An unknown error has occurred in the Tree class function assembleTree. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; - exit(1); - } } /*****************************************************************/ void Tree::getCopy(Tree* copy) { @@ -132,15 +214,12 @@ void Tree::getCopy(Tree* copy) { //copy pcount tree[i].pcount = copy->tree[i].pcount; } + } catch(exception& e) { - cout << "Standard Error: " << e.what() << " has occurred in the Tree class Function getCopy. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + m->errorOut(e, "Tree", "getCopy"); exit(1); } - catch(...) { - cout << "An unknown error has occurred in the Tree class function getCopy. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; - exit(1); - } } /*****************************************************************/ //returns a map with a groupname and the number of times that group was seen in the children @@ -175,11 +254,10 @@ map Tree::mergeGroups(int i) { // this is true if right child had a greater parsimony for a certain group if(maxPars > 1){ //erase all the groups that are only 1 because you found something with 2. - for(it=parsimony.begin();it!=parsimony.end();it++){ + for(it=parsimony.begin();it!=parsimony.end();){ if(it->second == 1){ - parsimony.erase(it->first); -// it--; - } + parsimony.erase(it++); + }else { it++; } } //set one remaining groups to 1 //so with our above example p[white] = 2 would be left and it would become p[white] = 1 @@ -192,13 +270,9 @@ map Tree::mergeGroups(int i) { return parsimony; } catch(exception& e) { - cout << "Standard Error: " << e.what() << " has occurred in the Tree class Function mergeGroups. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + m->errorOut(e, "Tree", "mergeGroups"); exit(1); } - catch(...) { - cout << "An unknown error has occurred in the Tree class function mergeGroups. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; - exit(1); - } } /*****************************************************************/ //returns a map with a groupname and the number of times that group was seen in the children @@ -213,13 +287,17 @@ map Tree::mergeUserGroups(int i, vector g) { int rc = tree[i].getRChild(); //loop through nodes groups removing the ones the user doesn't want - for (it = tree[lc].pGroups.begin(); it != tree[lc].pGroups.end(); it++) { - if (inUsersGroups(it->first, g) != true) { tree[lc].pGroups.erase(it->first); } + for(it=tree[lc].pGroups.begin();it!=tree[lc].pGroups.end();){ + if (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(); it++) { - if (inUsersGroups(it->first, g) != true) { tree[rc].pGroups.erase(it->first); } + for(it=tree[rc].pGroups.begin();it!=tree[rc].pGroups.end();){ + if (inUsersGroups(it->first, g) != true) { + tree[rc].pGroups.erase(it++); + }else { it++; } } //set parsimony groups to left child @@ -244,11 +322,12 @@ map Tree::mergeUserGroups(int i, vector g) { // this is true if right child had a greater parsimony for a certain group if(maxPars > 1){ //erase all the groups that are only 1 because you found something with 2. - for(it=parsimony.begin();it!=parsimony.end();it++){ + for(it=parsimony.begin();it!=parsimony.end();){ if(it->second == 1){ - parsimony.erase(it->first); - } + parsimony.erase(it++); + }else { it++; } } + for(it=parsimony.begin();it!=parsimony.end();it++){ parsimony[it->first] = 1; } @@ -257,13 +336,9 @@ map Tree::mergeUserGroups(int i, vector g) { return parsimony; } catch(exception& e) { - cout << "Standard Error: " << e.what() << " has occurred in the Tree class Function mergeGroups. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + m->errorOut(e, "Tree", "mergeUserGroups"); exit(1); } - catch(...) { - cout << "An unknown error has occurred in the Tree class function mergeGroups. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; - exit(1); - } } @@ -284,13 +359,9 @@ map Tree::mergeGcounts(int position) { return sum; } catch(exception& e) { - cout << "Standard Error: " << e.what() << " has occurred in the Tree class Function mergeGcounts. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + m->errorOut(e, "Tree", "mergeGcounts"); exit(1); } - catch(...) { - cout << "An unknown error has occurred in the Tree class function mergeGcounts. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; - exit(1); - } } /**************************************************************************************************/ @@ -315,7 +386,7 @@ void Tree::randomLabels(vector g) { tree[z].pGroups = (tree[i].pGroups); tree[i].pGroups = (lib_hold); - string zgroup = tree[z].getGroup(); + vector zgroup = tree[z].getGroup(); tree[z].setGroup(tree[i].getGroup()); tree[i].setGroup(zgroup); @@ -330,15 +401,11 @@ void Tree::randomLabels(vector g) { } } catch(exception& e) { - cout << "Standard Error: " << e.what() << " has occurred in the Tree class Function randomLabels. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + m->errorOut(e, "Tree", "randomLabels"); exit(1); } - catch(...) { - cout << "An unknown error has occurred in the Tree class function randomLabels. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; - exit(1); - } } -/**************************************************************************************************/ +/************************************************************************************************** void Tree::randomLabels(string groupA, string groupB) { try { @@ -364,13 +431,9 @@ void Tree::randomLabels(string groupA, string groupB) { } } catch(exception& e) { - cout << "Standard Error: " << e.what() << " has occurred in the Tree class Function randomLabels. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + m->errorOut(e, "Tree", "randomLabels"); exit(1); } - catch(...) { - cout << "An unknown error has occurred in the Tree class function randomLabels. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; - exit(1); - } } /**************************************************************************************************/ void Tree::randomBlengths() { @@ -384,13 +447,9 @@ void Tree::randomBlengths() { } } catch(exception& e) { - cout << "Standard Error: " << e.what() << " has occurred in the Tree class Function randomBlengths. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + m->errorOut(e, "Tree", "randomBlengths"); exit(1); } - catch(...) { - cout << "An unknown error has occurred in the Tree class function randomBlengths. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; - exit(1); - } } /*************************************************************************************************/ void Tree::assembleRandomUnifracTree(vector g) { @@ -399,7 +458,9 @@ void Tree::assembleRandomUnifracTree(vector g) { } /*************************************************************************************************/ void Tree::assembleRandomUnifracTree(string groupA, string groupB) { - randomLabels(groupA, groupB); + + vector temp; temp.push_back(groupA); temp.push_back(groupB); + randomLabels(temp); assembleTree(); } @@ -435,7 +496,7 @@ void Tree::randomTopology() { escape = 1; } } - + tree[i].setChildren(rnd_index1,rnd_index2); tree[i].setParent(-1); tree[rnd_index1].setParent(i); @@ -443,13 +504,33 @@ void Tree::randomTopology() { } } catch(exception& e) { - cout << "Standard Error: " << e.what() << " has occurred in the Tree class Function randomTopology. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + m->errorOut(e, "Tree", "randomTopology"); exit(1); } - catch(...) { - cout << "An unknown error has occurred in the Tree class function randomTopology. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; +} +/*****************************************************************/ +void Tree::print(ostream& out) { + try { + int root = findRoot(); + printBranch(root, out, "branch"); + out << ";" << endl; + } + catch(exception& e) { + m->errorOut(e, "Tree", "print"); exit(1); - } + } +} +/*****************************************************************/ +void Tree::printForBoot(ostream& out) { + try { + int root = findRoot(); + printBranch(root, out, "boot"); + out << ";" << endl; + } + catch(exception& e) { + m->errorOut(e, "Tree", "printForBoot"); + exit(1); + } } /*****************************************************************/ @@ -459,22 +540,19 @@ void Tree::createNewickFile(string f) { int root = findRoot(); //filename = getRootName(globaldata->getTreeFile()) + "newick"; filename = f; + openOutputFile(filename, out); - printBranch(root); + printBranch(root, out, "branch"); // you are at the end of the tree out << ";" << endl; out.close(); } catch(exception& e) { - cout << "Standard Error: " << e.what() << " has occurred in the Tree class Function createNewickFile. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + m->errorOut(e, "Tree", "createNewickFile"); exit(1); } - catch(...) { - cout << "An unknown error has occurred in the Tree class function createNewickFile. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; - exit(1); - } } /*****************************************************************/ @@ -485,43 +563,61 @@ int Tree::findRoot() { for (int i = 0; i < numNodes; i++) { //you found the root if (tree[i].getParent() == -1) { return i; } + //cout << "i = " << i << endl; + //cout << "i's parent = " << tree[i].getParent() << endl; } return -1; } catch(exception& e) { - cout << "Standard Error: " << e.what() << " has occurred in the Tree class Function findRoot. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + m->errorOut(e, "Tree", "findRoot"); exit(1); } - catch(...) { - cout << "An unknown error has occurred in the Tree class function findRoot. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; - exit(1); - } } /*****************************************************************/ -void Tree::printBranch(int node) { +void Tree::printBranch(int node, ostream& out, string mode) { try { // you are not a leaf if (tree[node].getLChild() != -1) { out << "("; - printBranch(tree[node].getLChild()); + printBranch(tree[node].getLChild(), out, mode); out << ","; - printBranch(tree[node].getRChild()); + printBranch(tree[node].getRChild(), out, mode); out << ")"; + if (mode == "branch") { + //if there is a branch length then print it + if (tree[node].getBranchLength() != -1) { + out << ":" << tree[node].getBranchLength(); + } + }else if (mode == "boot") { + //if there is a label then print it + if (tree[node].getLabel() != -1) { + out << tree[node].getLabel(); + } + } }else { //you are a leaf - out << tree[node].getGroup() << ":" << tree[node].getBranchLength(); + string leafGroup = globaldata->gTreemap->getGroup(tree[node].getName()); + + out << leafGroup; + if (mode == "branch") { + //if there is a branch length then print it + if (tree[node].getBranchLength() != -1) { + out << ":" << tree[node].getBranchLength(); + } + }else if (mode == "boot") { + //if there is a label then print it + if (tree[node].getLabel() != -1) { + out << tree[node].getLabel(); + } + } } } catch(exception& e) { - cout << "Standard Error: " << e.what() << " has occurred in the Tree class Function printBranch. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + m->errorOut(e, "Tree", "printBranch"); exit(1); } - catch(...) { - cout << "An unknown error has occurred in the Tree class function printBranch. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; - exit(1); - } } /*****************************************************************/ @@ -536,5 +632,155 @@ void Tree::printTree() { } /*****************************************************************/ +//this code is a mess and should be rethought...-slw +void Tree::parseTreeFile() { + + //only takes names from the first tree and assumes that all trees use the same names. + try { + string filename = globaldata->getTreeFile(); + ifstream filehandle; + openInputFile(filename, filehandle); + int c, comment; + comment = 0; + int done = 1; + + //ifyou are not a nexus file + if((c = filehandle.peek()) != '#') { + while((c = filehandle.peek()) != ';') { + while ((c = filehandle.peek()) != ';') { + // get past comments + if(c == '[') { + comment = 1; + } + if(c == ']'){ + comment = 0; + } + if((c == '(') && (comment != 1)){ break; } + filehandle.get(); + } + + done = readTreeString(filehandle); + if (done == 0) { break; } + } + //ifyou are a nexus file + }else if((c = filehandle.peek()) == '#') { + string holder = ""; + + // get past comments + while(holder != "translate" && holder != "Translate"){ + if(holder == "[" || holder == "[!"){ + comment = 1; + } + if(holder == "]"){ + comment = 0; + } + filehandle >> holder; + + //if there is no translate then you must read tree string otherwise use translate to get names + if((holder == "tree") && (comment != 1)){ + //pass over the "tree rep.6878900 = " + while (((c = filehandle.get()) != '(') && ((c = filehandle.peek()) != EOF)) {;} + + if(c == EOF) { break; } + filehandle.putback(c); //put back first ( of tree. + done = readTreeString(filehandle); + + break; + } + + if (done == 0) { break; } + } + + //use nexus translation rather than parsing tree to save time + if((holder == "translate") || (holder == "Translate")) { + + string number, name, h; + h = ""; // so it enters the loop the first time + while((h != ";") && (number != ";")) { + filehandle >> number; + filehandle >> name; + + //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); + } + if(number == ";") { globaldata->Treenames.pop_back(); } //in case ';' from translation is on next line instead of next to last name + } + } + filehandle.close(); + } + catch(exception& e) { + m->errorOut(e, "Tree", "parseTreeFile"); + exit(1); + } +} +/*******************************************************/ + +/*******************************************************/ +int Tree::readTreeString(ifstream& filehandle) { + try { + int c; + string name; //, k + + while((c = filehandle.peek()) != ';') { +//k = c; +//cout << " at beginning of while " << k << endl; + if(c == ')') { + //to pass over labels in trees + 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 = ""; + 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 + 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(); +//k = c; + //cout << " here after get " << k << endl; + if(c == ';') { return 0; } + if(c == ')') { filehandle.putback(c); } + //k = c; +//cout << k << endl; + + } + return 0; + } + catch(exception& e) { + m->errorOut(e, "Tree", "readTreeString"); + exit(1); + } +} + +/*******************************************************/ +/*******************************************************/