From 825512fb96e5bb00e08665396d9fa6a8c4da5a68 Mon Sep 17 00:00:00 2001 From: westcott Date: Wed, 27 May 2009 23:17:57 +0000 Subject: [PATCH] fixed multiple bugs for 1.4 release modified tree.shared to allow using a distance file --- engine.cpp | 3 + errorchecking.cpp | 25 ++- getgroupcommand.cpp | 48 +++++- getgroupcommand.h | 7 +- globaldata.cpp | 2 + helpcommand.cpp | 3 +- sharedchao1.cpp | 3 +- summarysharedcommand.cpp | 48 ++++-- summarysharedcommand.h | 5 +- tree.cpp | 4 +- tree.h | 2 +- treegroupscommand.cpp | 341 ++++++++++++++++++++++++++------------- treegroupscommand.h | 15 +- treemap.cpp | 35 +++- treemap.h | 4 +- validparameter.cpp | 2 +- 16 files changed, 386 insertions(+), 161 deletions(-) diff --git a/engine.cpp b/engine.cpp index ac49206..11d738d 100644 --- a/engine.cpp +++ b/engine.cpp @@ -127,6 +127,9 @@ BatchEngine::~BatchEngine(){ //This Function allows the user to run a batchfile containing several commands on Dotur bool BatchEngine::getInput(){ try { + //check if this is a valid batchfile + if (openedBatch == 1) { cout << "unable to open batchfile" << endl; return 1; } + string input = ""; string commandName = ""; bool errorFree; diff --git a/errorchecking.cpp b/errorchecking.cpp index 8b80580..e4c1877 100644 --- a/errorchecking.cpp +++ b/errorchecking.cpp @@ -236,7 +236,7 @@ bool ErrorCheck::checkInput(string input) { //check for valid method if(commandName == "get.group") { - if ((globaldata->getGroupFile() == "")) { cout << "You must read a group before you can use the get.group command." << endl; return false; } + if ((globaldata->getSharedFile() == "")) { cout << "You must read a groupfile or a sharedfile before you can use the get.group command." << endl; return false; } } if (commandName == "get.label" || commandName == "get.line") { if ((globaldata->getListFile() == "") && (globaldata->getRabundFile() == "") && (globaldata->getSabundFile() == "")) { cout << "You must read a list, sabund or rabund before you can use the get.label or get.line command." << endl; return false; } @@ -250,13 +250,22 @@ bool ErrorCheck::checkInput(string input) { if ((globaldata->getListFile() == "") && (globaldata->getRabundFile() == "") && (globaldata->getSabundFile() == "")) { cout << "You must read a list, sabund or rabund before you can use the collect.single, rarefaction.single or summary.single commands." << endl; return false; } } - if ((commandName == "collect.shared") || (commandName == "rarefaction.shared") || (commandName == "summary.shared") || (commandName == "tree.shared") || (commandName == "bootstrap.shared") || (commandName == "dist.shared")){ + if ((commandName == "collect.shared") || (commandName == "rarefaction.shared") || (commandName == "summary.shared") || (commandName == "bootstrap.shared") || (commandName == "dist.shared")){ if (globaldata->getSharedFile() == "") { if (globaldata->getListFile() == "") { cout << "You must read a list and a group, or a shared before you can use the collect.shared, rarefaction.shared, summary.shared, tree.shared, bootstrap.shared or dist.shared commands." << endl; return false; } else if (globaldata->getGroupFile() == "") { cout << "You must read a list and a group, or a shared before you can use the collect.shared, rarefaction.shared, summary.shared, tree.shared, bootstrap.shared or dist.shared commands." << endl; return false; } } } + if (commandName == "tree.shared") { + //given no files + if ((globaldata->getSharedFile() == "") && ((phylipfile == "") && (columnfile == ""))) { cout << "You must run the read.otu command or provide a distance file before running the tree.shared command." << endl; return false; } + //you want to do single commands + else if ((globaldata->getSharedFile() == "") && ((phylipfile != "") || (columnfile != ""))) { + validateReadDist(); + } + } + if ((commandName == "heatmap") || (commandName == "venn")) { if ((globaldata->getListFile() == "") && (globaldata->getSharedFile() == "")) { cout << "You must read a list, or a list and a group, or a shared before you can use the heatmap or venn commands." << endl; return false; @@ -354,6 +363,12 @@ void ErrorCheck::validateReadFiles() { //unable to open if (ableToOpen == 1) { errorFree = false; } else { globaldata->inputFileName = sharedfile; } + }else if (groupfile != "") { + ableToOpen = openInputFile(groupfile, filehandle); + filehandle.close(); + if (ableToOpen == 1) { //unable to open + errorFree = false; + } }else{ //no file given errorFree = false; } @@ -385,8 +400,8 @@ void ErrorCheck::validateReadDist() { if (ableToOpen == 1) { errorFree = false; } } - if ((phylipfile == "") && (columnfile == "")) { cout << "When executing a read.dist you must enter a phylip or a column." << endl; errorFree = false; } - else if ((phylipfile != "") && (columnfile != "")) { cout << "When executing a read.dist you must enter ONLY ONE of the following: phylip or column." << endl; errorFree = false; } + if ((phylipfile == "") && (columnfile == "")) { cout << "When executing a read.dist or a tree.shared command with a distance file you must enter a phylip or a column." << endl; errorFree = false; } + else if ((phylipfile != "") && (columnfile != "")) { cout << "When executing a read.dist or a tree.shared command with a distance file you must enter ONLY ONE of the following: phylip or column." << endl; errorFree = false; } if (columnfile != "") { if (namefile == "") { @@ -435,7 +450,7 @@ void ErrorCheck::validateParseFiles() { ableToOpen = openInputFile(groupfile, filehandle); filehandle.close(); if (ableToOpen == 1) { //unable to open - errorFree = false;; + errorFree = false; } } } diff --git a/getgroupcommand.cpp b/getgroupcommand.cpp index 3fc67d1..4d43959 100644 --- a/getgroupcommand.cpp +++ b/getgroupcommand.cpp @@ -13,8 +13,13 @@ GetgroupCommand::GetgroupCommand(){ try { globaldata = GlobalData::getInstance(); - groupMap = globaldata->gGroupmap; - outputFile = globaldata->inputFileName + ".bootGroups"; + + //open shared file + sharedfile = globaldata->getSharedFile(); + openInputFile(sharedfile, in); + + //open output file + outputFile = getRootName(globaldata->inputFileName) + "bootGroups"; openOutputFile(outputFile, out); } catch(exception& e) { @@ -37,11 +42,42 @@ GetgroupCommand::~GetgroupCommand(){ int GetgroupCommand::execute(){ try { - vector groupNames = groupMap->namesOfGroups; - for(int i = 0; i < groupNames.size(); i++) { - cout << groupNames[i] << "\n"; - out << groupNames[i] << "\t" << groupNames[i] << "\n"; + int num, inputData, count; + count = 0; + string holdLabel, nextLabel, groupN, label; + + //read in first row since you know there is at least 1 group. + in >> label >> groupN >> num; + holdLabel = label; + + //output first group + cout << groupN << endl; + out << groupN << '\t' << groupN << endl; + + //get rest of line + for(int i=0;i> inputData; + } + + if (in.eof() != true) { in >> nextLabel; } + + //read the rest of the groups info in + while ((nextLabel == holdLabel) && (in.eof() != true)) { + in >> groupN >> num; + count++; + + //output next group + cout << groupN << endl; + out << groupN << '\t' << groupN << endl; + + //fill vector. + for(int i=0;i> inputData; + } + + if (in.eof() != true) { in >> nextLabel; } } + out.close(); return 0; } diff --git a/getgroupcommand.h b/getgroupcommand.h index 26190b5..496d0c9 100644 --- a/getgroupcommand.h +++ b/getgroupcommand.h @@ -11,9 +11,7 @@ */ #include "command.hpp" -#include "readmatrix.hpp" - -class GlobalData; +#include "globaldata.hpp" class GetgroupCommand : public Command { public: @@ -24,8 +22,9 @@ public: private: GlobalData* globaldata; GroupMap* groupMap; - string outputFile; + string outputFile, sharedfile; ofstream out; + ifstream in; }; diff --git a/globaldata.cpp b/globaldata.cpp index 1be2c6e..02b9013 100644 --- a/globaldata.cpp +++ b/globaldata.cpp @@ -249,10 +249,12 @@ void GlobalData::parseGlobalData(string commandString, string optionText){ Estimators.clear(); splitAtDash(calc, Estimators); } + if(commandName == "filter.seqs"){ if(trump == "" && vertical == "" && hard == "" && soft == ""){ trump = '.'; } + } //if you have done a read.otu with a groupfile but don't want to use it anymore because you want to do single commands diff --git a/helpcommand.cpp b/helpcommand.cpp index b480dcb..731d77e 100644 --- a/helpcommand.cpp +++ b/helpcommand.cpp @@ -121,7 +121,8 @@ int HelpCommand::execute(){ cout << "The groups parameter allows you to specify which of the groups in your groupfile you would like analyzed. You must enter at least 2 valid groups." << "\n"; cout << "Note: No spaces between parameter labels (i.e. list), '=' and parameters (i.e.yourListfile)." << "\n" << "\n"; }else if (globaldata->helpRequest == "get.group") { - cout << "The get.group command can only be executed after a successful read.otu command of a group file." << "\n"; + cout << "The get.group command can only be executed after a successful read.otu command." << "\n"; +//cout << "The get.group command outputs a .bootGroups file to you can use in addition to the tree file generated by the bootstrap.shared command to run the concensus command." << "\n"; cout << "You may not use any parameters with the get.group command." << "\n"; cout << "The get.group command should be in the following format: " << "\n"; cout << "get.group()" << "\n"; diff --git a/sharedchao1.cpp b/sharedchao1.cpp index 7c617c5..8dff314 100644 --- a/sharedchao1.cpp +++ b/sharedchao1.cpp @@ -12,8 +12,7 @@ /***********************************************************************/ EstOutput SharedChao1::getValues(vector shared){ try { - data.resize(1,0); - + data.resize(1,0); vector temp; int numGroups = shared.size(); float Chao = 0.0; float leftvalue, rightvalue; diff --git a/summarysharedcommand.cpp b/summarysharedcommand.cpp index 46055c0..d1d1be0 100644 --- a/summarysharedcommand.cpp +++ b/summarysharedcommand.cpp @@ -42,7 +42,6 @@ SummarySharedCommand::SummarySharedCommand(){ openOutputFile(outputFileName, outputFileHandle); format = globaldata->getFormat(); validCalculator = new ValidCalculators(); - util = new SharedUtil(); mult = false; int i; @@ -111,7 +110,6 @@ SummarySharedCommand::SummarySharedCommand(){ SummarySharedCommand::~SummarySharedCommand(){ delete input; delete read; - delete util; } //********************************************************************************************************************** @@ -119,7 +117,7 @@ SummarySharedCommand::~SummarySharedCommand(){ int SummarySharedCommand::execute(){ try { int count = 1; - + //if the users entered no valid calculators don't execute command if (sumCalculators.size() == 0) { return 0; } //check if any calcs can do multiples @@ -129,14 +127,12 @@ int SummarySharedCommand::execute(){ } } + //read first line read = new ReadOTUFile(globaldata->inputFileName); read->read(&*globaldata); input = globaldata->ginput; - order = input->getSharedOrderVector(); - - //set users groups - util->setGroups(globaldata->Groups, globaldata->gGroupmap->namesOfGroups, "summary"); + lookup = input->getSharedRAbundVectors(); //output estimator names as column headers outputFileHandle << "label" <<'\t' << "comparison" << '\t'; @@ -146,7 +142,7 @@ int SummarySharedCommand::execute(){ outputFileHandle << endl; //create file and put column headers for multiple groups file - if (mult = true) { + if (mult == true) { outAllFileName = ((getRootName(globaldata->inputFileName)) + "sharedmultiple.summary"); openOutputFile(outAllFileName, outAll); @@ -159,17 +155,27 @@ int SummarySharedCommand::execute(){ outAll << endl; } - while(order != NULL){ + if (lookup.size() < 2) { + cout << "I cannot run the command without at least 2 valid groups."; + for (int i = 0; i < lookup.size(); i++) { delete lookup[i]; } + + //close files and clean up + outputFileHandle.close(); remove(outputFileName.c_str()); + if (mult == true) { outAll.close(); remove(outAllFileName.c_str()); } + return 0; + } + + + while(lookup[0] != NULL){ - if(globaldata->allLines == 1 || globaldata->lines.count(count) == 1 || globaldata->labels.count(order->getLabel()) == 1){ + if(globaldata->allLines == 1 || globaldata->lines.count(count) == 1 || globaldata->labels.count(lookup[0]->getLabel()) == 1){ - cout << order->getLabel() << '\t' << count << endl; - util->getSharedVectors(globaldata->Groups, lookup, order); //fills group vectors from order vector. //fills group vectors from order vector. + cout << lookup[0]->getLabel() << '\t' << count << endl; //loop through calculators and add to file all for all calcs that can do mutiple groups - if (mult = true) { + if (mult == true) { //output label - outAll << order->getLabel() << '\t'; + outAll << lookup[0]->getLabel() << '\t'; //output groups names string outNames = ""; @@ -188,13 +194,13 @@ int SummarySharedCommand::execute(){ } outAll << endl; } - + int n = 1; vector subset; for (int k = 0; k < (lookup.size() - 1); k++) { // pass cdd each set of groups to commpare for (int l = n; l < lookup.size(); l++) { - outputFileHandle << order->getLabel() << '\t'; + outputFileHandle << lookup[0]->getLabel() << '\t'; subset.clear(); //clear out old pair of sharedrabunds //add new pair of sharedrabunds @@ -208,6 +214,7 @@ int SummarySharedCommand::execute(){ } for(int i=0;igetValues(subset); //saves the calculator outputs outputFileHandle << '\t'; sumCalculators[i]->print(outputFileHandle); @@ -218,13 +225,20 @@ int SummarySharedCommand::execute(){ } } + //prevent memory leak + for (int i = 0; i < lookup.size(); i++) { delete lookup[i]; } + //get next line to process - order = input->getSharedOrderVector(); + lookup = input->getSharedRAbundVectors(); count++; } //reset groups parameter globaldata->Groups.clear(); globaldata->setGroups(""); + + //close files + outputFileHandle.close(); + if (mult == true) { outAll.close(); } return 0; } diff --git a/summarysharedcommand.h b/summarysharedcommand.h index ea141d1..e73714a 100644 --- a/summarysharedcommand.h +++ b/summarysharedcommand.h @@ -15,9 +15,8 @@ #include "inputdata.h" #include "calculator.h" #include "readotu.h" -#include "sharedlistvector.h" #include "validcalculator.h" -#include "sharedutilities.h" + /*The summary.shared() command The summary.shared command can only be executed after a successful read.shared command. @@ -46,12 +45,10 @@ public: private: GlobalData* globaldata; ReadOTUFile* read; - SharedUtil* util; vector sumCalculators; InputData* input; ValidCalculators* validCalculator; SharedListVector* SharedList; - SharedOrderVector* order; vector lookup; string outputFileName, format, outAllFileName; ofstream outputFileHandle, outAll; diff --git a/tree.cpp b/tree.cpp index a6c9c07..30df409 100644 --- a/tree.cpp +++ b/tree.cpp @@ -518,7 +518,9 @@ int Tree::findRoot() { try { for (int i = 0; i < numNodes; i++) { //you found the root - if (tree[i].getParent() == -1) { return i; } + if (tree[i].getParent() == -1) { return i; } + //cout << "i = " << i << endl; + //cout << "i's parent = " << tree[i].getParent() << endl; } return -1; } diff --git a/tree.h b/tree.h index 8b77d3d..7fc3838 100644 --- a/tree.h +++ b/tree.h @@ -36,6 +36,7 @@ public: void printTree(); void print(ostream&); void printForBoot(ostream&); + int findRoot(); //return index of root node //this function takes the leaf info and populates the non leaf nodes void assembleTree(); @@ -55,7 +56,6 @@ private: void randomBlengths(); void randomLabels(vector); void randomLabels(string, string); - int findRoot(); //return index of root node void printBranch(int, ostream&, string); //recursively print out tree }; diff --git a/treegroupscommand.cpp b/treegroupscommand.cpp index 43e39f5..a21c47a 100644 --- a/treegroupscommand.cpp +++ b/treegroupscommand.cpp @@ -27,30 +27,32 @@ TreeGroupCommand::TreeGroupCommand(){ globaldata = GlobalData::getInstance(); format = globaldata->getFormat(); validCalculator = new ValidCalculators(); - - int i; - for (i=0; iEstimators.size(); i++) { - if (validCalculator->isValidCalculator("treegroup", globaldata->Estimators[i]) == true) { - if (globaldata->Estimators[i] == "jabund") { - treeCalculators.push_back(new JAbund()); - }else if (globaldata->Estimators[i] == "sorabund") { - treeCalculators.push_back(new SorAbund()); - }else if (globaldata->Estimators[i] == "jclass") { - treeCalculators.push_back(new Jclass()); - }else if (globaldata->Estimators[i] == "sorclass") { - treeCalculators.push_back(new SorClass()); - }else if (globaldata->Estimators[i] == "jest") { - treeCalculators.push_back(new Jest()); - }else if (globaldata->Estimators[i] == "sorest") { - treeCalculators.push_back(new SorEst()); - }else if (globaldata->Estimators[i] == "thetayc") { - treeCalculators.push_back(new ThetaYC()); - }else if (globaldata->Estimators[i] == "thetan") { - treeCalculators.push_back(new ThetaN()); - }else if (globaldata->Estimators[i] == "morisitahorn") { - treeCalculators.push_back(new MorHorn()); - }else if (globaldata->Estimators[i] == "braycurtis") { - treeCalculators.push_back(new BrayCurtis()); + + if (format == "sharedfile") { + int i; + for (i=0; iEstimators.size(); i++) { + if (validCalculator->isValidCalculator("treegroup", globaldata->Estimators[i]) == true) { + if (globaldata->Estimators[i] == "jabund") { + treeCalculators.push_back(new JAbund()); + }else if (globaldata->Estimators[i] == "sorabund") { + treeCalculators.push_back(new SorAbund()); + }else if (globaldata->Estimators[i] == "jclass") { + treeCalculators.push_back(new Jclass()); + }else if (globaldata->Estimators[i] == "sorclass") { + treeCalculators.push_back(new SorClass()); + }else if (globaldata->Estimators[i] == "jest") { + treeCalculators.push_back(new Jest()); + }else if (globaldata->Estimators[i] == "sorest") { + treeCalculators.push_back(new SorEst()); + }else if (globaldata->Estimators[i] == "thetayc") { + treeCalculators.push_back(new ThetaYC()); + }else if (globaldata->Estimators[i] == "thetan") { + treeCalculators.push_back(new ThetaN()); + }else if (globaldata->Estimators[i] == "morisitahorn") { + treeCalculators.push_back(new MorHorn()); + }else if (globaldata->Estimators[i] == "braycurtis") { + treeCalculators.push_back(new BrayCurtis()); + } } } } @@ -72,100 +74,81 @@ TreeGroupCommand::TreeGroupCommand(){ TreeGroupCommand::~TreeGroupCommand(){ delete input; - delete read; + if (format == "sharedfile") {delete read;} + else { delete readMatrix; delete matrix; delete list; } + delete tmap; + } //********************************************************************************************************************** int TreeGroupCommand::execute(){ try { - int count = 1; - EstOutput data; - vector subset; - - //if the users entered no valid calculators don't execute command - if (treeCalculators.size() == 0) { return 0; } + if (format == "sharedfile") { + //if the users entered no valid calculators don't execute command + if (treeCalculators.size() == 0) { cout << "You have given no valid calculators." << endl; return 0; } - //you have groups - read = new ReadOTUFile(globaldata->inputFileName); - read->read(&*globaldata); + //you have groups + read = new ReadOTUFile(globaldata->inputFileName); + read->read(&*globaldata); - input = globaldata->ginput; - lookup = input->getSharedRAbundVectors(); - - if (lookup.size() < 2) { cout << "You have not provided enough valid groups. I cannot run the command." << endl; return 0; } - - numGroups = globaldata->Groups.size(); - groupNames = ""; - for (int i = 0; i < numGroups; i++) { groupNames += globaldata->Groups[i]; } - - //clear globaldatas old tree names if any - globaldata->Treenames.clear(); - - //fills globaldatas tree names - globaldata->Treenames = globaldata->Groups; - - //create treemap class from groupmap for tree class to use - tmap = new TreeMap(); - tmap->makeSim(globaldata->gGroupmap); - globaldata->gTreemap = tmap; + input = globaldata->ginput; + lookup = input->getSharedRAbundVectors(); - while(lookup[0] != NULL){ + if (lookup.size() < 2) { cout << "You have not provided enough valid groups. I cannot run the command." << endl; return 0; } - if(globaldata->allLines == 1 || globaldata->lines.count(count) == 1 || globaldata->labels.count(lookup[0]->getLabel()) == 1){ - - cout << lookup[0]->getLabel() << '\t' << count << endl; - - //for each calculator - for(int i = 0 ; i < treeCalculators.size(); i++) { - - //initialize simMatrix - simMatrix.clear(); - simMatrix.resize(numGroups); - for (int m = 0; m < simMatrix.size(); m++) { - for (int j = 0; j < simMatrix.size(); j++) { - simMatrix[m].push_back(0.0); - } - } + //create tree file + makeSimsShared(); + }else{ + //read in dist file + filename = globaldata->inputFileName; + + if (format == "column") { readMatrix = new ReadColumnMatrix(filename); } + else if (format == "phylip") { readMatrix = new ReadPhylipMatrix(filename); } - //initialize index - index.clear(); - for (int g = 0; g < numGroups; g++) { index[g] = g; } + if(globaldata->getPrecision() != ""){ + convert(globaldata->getPrecision(), precision); + } - //create a new filename - outputFile = getRootName(globaldata->inputFileName) + treeCalculators[i]->getName() + "." + lookup[0]->getLabel() + ".tre"; - - - for (int k = 0; k < lookup.size(); k++) { - for (int l = k; l < lookup.size(); l++) { - if (k != l) { //we dont need to similiarity of a groups to itself - //get estimated similarity between 2 groups - - subset.clear(); //clear out old pair of sharedrabunds - //add new pair of sharedrabunds - subset.push_back(lookup[k]); subset.push_back(lookup[l]); - - data = treeCalculators[i]->getValues(subset); //saves the calculator outputs - //save values in similarity matrix - simMatrix[k][l] = data[0]; - simMatrix[l][k] = data[0]; - } - } - } - - //creates tree from similarity matrix and write out file - createTree(); - } + if(globaldata->getCutOff() != ""){ + convert(globaldata->getCutOff(), cutoff); + cutoff += (5 / (precision * 10.0)); } + readMatrix->setCutoff(cutoff); + + if(globaldata->getNameFile() != ""){ + nameMap = new NameAssignment(globaldata->getNameFile()); + nameMap->readMap(1,2); + } + else{ + nameMap = NULL; + } + + readMatrix->read(nameMap); + list = readMatrix->getListVector(); + matrix = readMatrix->getMatrix(); + + //make treemap + tmap = new TreeMap(); + tmap->makeSim(list); + globaldata->gTreemap = tmap; - //prevent memory leak - for (int i = 0; i < lookup.size(); i++) { delete lookup[i]; } - - //get next line to process - lookup = input->getSharedRAbundVectors(); - count++; - } + globaldata->Groups = tmap->namesOfGroups; + + //clear globaldatas old tree names if any + globaldata->Treenames.clear(); + //fills globaldatas tree names + globaldata->Treenames = globaldata->Groups; + + makeSimsDist(); + + //create a new filename + outputFile = getRootName(globaldata->inputFileName) + "tre"; + + createTree(); + } + //reset groups parameter globaldata->Groups.clear(); globaldata->setGroups(""); @@ -190,8 +173,8 @@ void TreeGroupCommand::createTree(){ //do merges and create tree structure by setting parents and children //there are numGroups - 1 merges to do for (int i = 0; i < (numGroups - 1); i++) { - - float largest = 0.0; + float largest = -1000.0; + int row, column; //find largest value in sims matrix by searching lower triangle for (int j = 1; j < simMatrix.size(); j++) { @@ -223,9 +206,9 @@ void TreeGroupCommand::createTree(){ index[row] = numGroups+i; index[column] = numGroups+i; - //zero out highest value that caused the merge. - simMatrix[row][column] = 0.0; - simMatrix[column][row] = 0.0; + //remove highest value that caused the merge. + simMatrix[row][column] = -1000.0; + simMatrix[column][row] = -1000.0; //merge values in simsMatrix for (int n = 0; n < simMatrix.size(); n++) { @@ -233,11 +216,15 @@ void TreeGroupCommand::createTree(){ simMatrix[row][n] = (simMatrix[row][n] + simMatrix[column][n]) / 2; simMatrix[n][row] = simMatrix[row][n]; //delete column - simMatrix[column][n] = 0.0; - simMatrix[n][column] = 0.0; + simMatrix[column][n] = -1000.0; + simMatrix[n][column] = -1000.0; } } - + + //adjust tree to make sure root to tip length is .5 + int root = t->findRoot(); + t->tree[root].setBranchLength((0.5 - t->tree[root].getLengthToLeaves())); + //assemble tree t->assembleTree(); @@ -262,13 +249,13 @@ void TreeGroupCommand::printSims(ostream& out) { try { //output column headers - out << '\t'; - for (int i = 0; i < lookup.size(); i++) { out << lookup[i]->getGroup() << '\t'; } - out << endl; + //out << '\t'; + //for (int i = 0; i < lookup.size(); i++) { out << lookup[i]->getGroup() << '\t'; } + //out << endl; for (int m = 0; m < simMatrix.size(); m++) { - out << lookup[m]->getGroup() << '\t'; + //out << lookup[m]->getGroup() << '\t'; for (int n = 0; n < simMatrix.size(); n++) { out << simMatrix[m][n] << '\t'; } @@ -286,5 +273,129 @@ void TreeGroupCommand::printSims(ostream& out) { } } /***********************************************************/ +void TreeGroupCommand::makeSimsDist() { + try { + numGroups = list->size(); + + //initialize index + index.clear(); + for (int g = 0; g < numGroups; g++) { index[g] = g; } + + //initialize simMatrix + simMatrix.clear(); + simMatrix.resize(numGroups); + for (int m = 0; m < simMatrix.size(); m++) { + for (int j = 0; j < simMatrix.size(); j++) { + simMatrix[m].push_back(0.0); + } + } + + //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); + } + + + } + catch(exception& e) { + cout << "Standard Error: " << e.what() << " has occurred in the TreeGroupCommand class Function makeSimsDist. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + exit(1); + } + catch(...) { + cout << "An unknown error has occurred in the TreeGroupCommand class function makeSimsDist. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + exit(1); + } +} + +/***********************************************************/ +void TreeGroupCommand::makeSimsShared() { + try { + int count = 1; + EstOutput data; + vector subset; + + numGroups = globaldata->Groups.size(); + + //clear globaldatas old tree names if any + globaldata->Treenames.clear(); + + //fills globaldatas tree names + globaldata->Treenames = globaldata->Groups; + + //create treemap class from groupmap for tree class to use + tmap = new TreeMap(); + tmap->makeSim(globaldata->gGroupmap); + globaldata->gTreemap = tmap; + + while(lookup[0] != NULL){ + + if(globaldata->allLines == 1 || globaldata->lines.count(count) == 1 || globaldata->labels.count(lookup[0]->getLabel()) == 1){ + + cout << lookup[0]->getLabel() << '\t' << count << endl; + + //for each calculator + for(int i = 0 ; i < treeCalculators.size(); i++) { + //initialize simMatrix + simMatrix.clear(); + simMatrix.resize(numGroups); + for (int m = 0; m < simMatrix.size(); m++) { + for (int j = 0; j < simMatrix.size(); j++) { + simMatrix[m].push_back(0.0); + } + } + + //initialize index + index.clear(); + for (int g = 0; g < numGroups; g++) { index[g] = g; } + + //create a new filename + outputFile = getRootName(globaldata->inputFileName) + treeCalculators[i]->getName() + "." + lookup[0]->getLabel() + ".tre"; + + for (int k = 0; k < lookup.size(); k++) { + for (int l = k; l < lookup.size(); l++) { + if (k != l) { //we dont need to similiarity of a groups to itself + //get estimated similarity between 2 groups + + subset.clear(); //clear out old pair of sharedrabunds + //add new pair of sharedrabunds + subset.push_back(lookup[k]); subset.push_back(lookup[l]); + + data = treeCalculators[i]->getValues(subset); //saves the calculator outputs + //save values in similarity matrix + simMatrix[k][l] = data[0]; + simMatrix[l][k] = data[0]; + } + } + } + + //creates tree from similarity matrix and write out file + createTree(); + } + } + + //prevent memory leak + for (int i = 0; i < lookup.size(); i++) { delete lookup[i]; } + + //get next line to process + lookup = input->getSharedRAbundVectors(); + count++; + } + + for(int i = 0 ; i < treeCalculators.size(); i++) { delete treeCalculators[i]; } + } + catch(exception& e) { + cout << "Standard Error: " << e.what() << " has occurred in the TreeGroupCommand class Function makeSimsShared. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + exit(1); + } + catch(...) { + cout << "An unknown error has occurred in the TreeGroupCommand class function makeSimsShared. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + exit(1); + } +} + +/***********************************************************/ diff --git a/treegroupscommand.h b/treegroupscommand.h index 10a2e31..3f145e9 100644 --- a/treegroupscommand.h +++ b/treegroupscommand.h @@ -17,6 +17,10 @@ #include "validcalculator.h" #include "tree.h" #include "treemap.h" +#include "readmatrix.hpp" +#include "readcolumn.h" +#include "readphylip.h" +#include "sparsematrix.hpp" /* This command create a tree file for each similarity calculator at distance level, using various calculators to find the similiarity between groups. @@ -25,6 +29,8 @@ class GlobalData; +typedef list::iterator MatData; + class TreeGroupCommand : public Command { public: @@ -35,9 +41,15 @@ public: private: void createTree(); void printSims(ostream&); + void makeSimsShared(); + void makeSimsDist(); GlobalData* globaldata; ReadOTUFile* read; + ReadMatrix* readMatrix; + SparseMatrix* matrix; + NameAssignment* nameMap; + ListVector* list; TreeMap* tmap; Tree* t; vector treeCalculators; @@ -46,9 +58,10 @@ private: InputData* input; ValidCalculators* validCalculator; vector lookup; - string format, outputFile, groupNames; + string format, outputFile, groupNames, filename; int numGroups; ofstream out; + float precision, cutoff; }; diff --git a/treemap.cpp b/treemap.cpp index 00ae0e6..8e3fd58 100644 --- a/treemap.cpp +++ b/treemap.cpp @@ -162,13 +162,44 @@ void TreeMap::makeSim(GroupMap* groupmap) { } catch(exception& e) { - cout << "Standard Error: " << e.what() << " has occurred in the TreeMap class Function make. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + cout << "Standard Error: " << e.what() << " has occurred in the TreeMap class Function makeSim. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; exit(1); } catch(...) { - cout << "An unknown error has occurred in the TreeMap class function make. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + cout << "An unknown error has occurred in the TreeMap class function makeSim. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; exit(1); } } /************************************************************/ +void TreeMap::makeSim(ListVector* list) { + try { + //set names of groups + namesOfGroups.clear(); + for(int i = 0; i < list->size(); i++) { + namesOfGroups.push_back(list->get(i)); + } + + //set names of seqs to names of groups + namesOfSeqs = namesOfGroups; + + // make map where key and value are both the group name since that what the tree.shared command wants + for (int i = 0; i < namesOfGroups.size(); i++) { + treemap[namesOfGroups[i]].groupname = namesOfGroups[i]; + seqsPerGroup[namesOfGroups[i]] = 1; + } + + numGroups = namesOfGroups.size(); + + } + catch(exception& e) { + cout << "Standard Error: " << e.what() << " has occurred in the TreeMap class Function makeSim. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + exit(1); + } + catch(...) { + cout << "An unknown error has occurred in the TreeMap class function makeSim. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + exit(1); + } +} + +/************************************************************/ diff --git a/treemap.h b/treemap.h index 73d02fc..8f744c8 100644 --- a/treemap.h +++ b/treemap.h @@ -11,6 +11,7 @@ #include "mothur.h" #include "groupmap.h" +#include "listvector.hpp" /* This class is used by the read.tree command to build the tree container. */ @@ -20,6 +21,7 @@ struct GroupIndex { }; class GroupMap; +class ListVector; class TreeMap { public: @@ -39,7 +41,7 @@ public: map treemap; //sequence name and void print(ostream&); void makeSim(GroupMap*); //takes groupmap info and fills treemap for use by tree.shared command. - + void makeSim(ListVector*); //takes listvector info and fills treemap for use by tree.shared command. private: ifstream fileHandle; diff --git a/validparameter.cpp b/validparameter.cpp index 650ef1f..4aad1c2 100644 --- a/validparameter.cpp +++ b/validparameter.cpp @@ -282,7 +282,7 @@ void ValidParameters::initCommandParameters() { string getOTURepArray[] = {"fasta","list","line","label","name", "group"}; commandParameters["get.oturep"] = addParameters(getOTURepArray, sizeof(getOTURepArray)/sizeof(string)); - string treeGroupsArray[] = {"line","label","calc","groups"}; + string treeGroupsArray[] = {"line","label","calc","groups", "phylip", "column", "name"}; commandParameters["tree.shared"] = addParameters(treeGroupsArray, sizeof(treeGroupsArray)/sizeof(string)); string bootstrapArray[] = {"line","label","calc","groups","iters"}; -- 2.39.2