X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=bootstrapsharedcommand.cpp;h=afc3cf1533475b6bf78b8127ba34086193612371;hb=a4f05f09c250323132a4b9427f41a3ededaec634;hp=2835061392732dfa402f626a3d339726a601435a;hpb=a8f5a612bba54ceb74e17efc027d3a7f5aa93c9a;p=mothur.git diff --git a/bootstrapsharedcommand.cpp b/bootstrapsharedcommand.cpp index 2835061..afc3cf1 100644 --- a/bootstrapsharedcommand.cpp +++ b/bootstrapsharedcommand.cpp @@ -19,91 +19,238 @@ #include "sharedmorisitahorn.h" #include "sharedbraycurtis.h" - //********************************************************************************************************************** - -BootSharedCommand::BootSharedCommand(){ +vector BootSharedCommand::getValidParameters(){ + try { + string AlignArray[] = {"label","calc","groups","iters","outputdir","inputdir"}; + vector myArray (AlignArray, AlignArray+(sizeof(AlignArray)/sizeof(string))); + return myArray; + } + catch(exception& e) { + m->errorOut(e, "BootSharedCommand", "getValidParameters"); + exit(1); + } +} +//********************************************************************************************************************** +BootSharedCommand::BootSharedCommand(){ + try { + abort = true; calledHelp = true; + vector tempOutNames; + outputTypes["tree"] = tempOutNames; + } + catch(exception& e) { + m->errorOut(e, "BootSharedCommand", "BootSharedCommand"); + exit(1); + } +} +//********************************************************************************************************************** +vector BootSharedCommand::getRequiredParameters(){ + try { + vector myArray; + return myArray; + } + catch(exception& e) { + m->errorOut(e, "BootSharedCommand", "getRequiredParameters"); + exit(1); + } +} +//********************************************************************************************************************** +vector BootSharedCommand::getRequiredFiles(){ + try { + string AlignArray[] = {"shared"}; + vector myArray (AlignArray, AlignArray+(sizeof(AlignArray)/sizeof(string))); + return myArray; + } + catch(exception& e) { + m->errorOut(e, "BootSharedCommand", "getRequiredFiles"); + exit(1); + } +} +//********************************************************************************************************************** +BootSharedCommand::BootSharedCommand(string option) { try { globaldata = GlobalData::getInstance(); - format = globaldata->getFormat(); - convert(globaldata->getIters(), iters); - validCalculator = new ValidCalculators(); - util = new SharedUtil(); + abort = false; calledHelp = false; + allLines = 1; + labels.clear(); + Groups.clear(); + Estimators.clear(); + //allow user to run help + if(option == "help") { help(); abort = true; calledHelp = true; } - int i; - for (i=0; iEstimators.size(); i++) { - if (validCalculator->isValidCalculator("boot", 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()); - } + else { + //valid paramters for this command + string Array[] = {"label","calc","groups","iters","outputdir","inputdir"}; + vector myArray (Array, Array+(sizeof(Array)/sizeof(string))); + + OptionParser parser(option); + map parameters = parser.getParameters(); + + ValidParameters validParameter; + + //check to make sure all parameters are valid for command + for (map::iterator it = parameters.begin(); it != parameters.end(); it++) { + if (validParameter.isValidParameter(it->first, myArray, it->second) != true) { abort = true; } } - } + + //initialize outputTypes + vector tempOutNames; + outputTypes["tree"] = tempOutNames; - ofstream* temp; - for (int i=0; i < treeCalculators.size(); i++) { - temp = new ofstream; - out.push_back(temp); + //if the user changes the output directory command factory will send this info to us in the output parameter + outputDir = validParameter.validFile(parameters, "outputdir", false); if (outputDir == "not found"){ + outputDir = ""; + outputDir += m->hasPath(globaldata->inputFileName); //if user entered a file with a path then preserve it + } + + + //make sure the user has already run the read.otu command + if (globaldata->getSharedFile() == "") { + if (globaldata->getListFile() == "") { m->mothurOut("You must read a list and a group, or a shared before you can use the bootstrap.shared command."); m->mothurOutEndLine(); abort = true; } + else if (globaldata->getGroupFile() == "") { m->mothurOut("You must read a list and a group, or a shared before you can use the bootstrap.shared command."); m->mothurOutEndLine(); abort = true; } + } + + //check for optional parameter and set defaults + // ...at some point should added some additional type checking... + label = validParameter.validFile(parameters, "label", false); + if (label == "not found") { label = ""; } + else { + if(label != "all") { m->splitAtDash(label, labels); allLines = 0; } + else { allLines = 1; } + } + + //if the user has not specified any labels use the ones from read.otu + if(label == "") { + allLines = globaldata->allLines; + labels = globaldata->labels; + } + + groups = validParameter.validFile(parameters, "groups", false); + if (groups == "not found") { groups = ""; } + else { + m->splitAtDash(groups, Groups); + globaldata->Groups = Groups; + } + + calc = validParameter.validFile(parameters, "calc", false); + if (calc == "not found") { calc = "jclass-thetayc"; } + else { + if (calc == "default") { calc = "jclass-thetayc"; } + } + m->splitAtDash(calc, Estimators); + + string temp; + temp = validParameter.validFile(parameters, "iters", false); if (temp == "not found") { temp = "1000"; } + convert(temp, iters); + + if (abort == false) { + + //used in tree constructor + globaldata->runParse = false; + + validCalculator = new ValidCalculators(); + + int i; + for (i=0; iisValidCalculator("boot", Estimators[i]) == true) { + if (Estimators[i] == "jabund") { + treeCalculators.push_back(new JAbund()); + }else if (Estimators[i] == "sorabund") { + treeCalculators.push_back(new SorAbund()); + }else if (Estimators[i] == "jclass") { + treeCalculators.push_back(new Jclass()); + }else if (Estimators[i] == "sorclass") { + treeCalculators.push_back(new SorClass()); + }else if (Estimators[i] == "jest") { + treeCalculators.push_back(new Jest()); + }else if (Estimators[i] == "sorest") { + treeCalculators.push_back(new SorEst()); + }else if (Estimators[i] == "thetayc") { + treeCalculators.push_back(new ThetaYC()); + }else if (Estimators[i] == "thetan") { + treeCalculators.push_back(new ThetaN()); + }else if (Estimators[i] == "morisitahorn") { + treeCalculators.push_back(new MorHorn()); + }else if (Estimators[i] == "braycurtis") { + treeCalculators.push_back(new BrayCurtis()); + } + } + } + + delete validCalculator; + + ofstream* tempo; + for (int i=0; i < treeCalculators.size(); i++) { + tempo = new ofstream; + out.push_back(tempo); + } + + //make a vector of tree* for each calculator + trees.resize(treeCalculators.size()); + } } - - //reset calc for next command - globaldata->setCalc(""); } catch(exception& e) { - cout << "Standard Error: " << e.what() << " has occurred in the BootSharedCommand class Function BootSharedCommand. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + m->errorOut(e, "BootSharedCommand", "BootSharedCommand"); exit(1); } - catch(...) { - cout << "An unknown error has occurred in the BootSharedCommand class function BootSharedCommand. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; +} + +//********************************************************************************************************************** + +void BootSharedCommand::help(){ + try { + m->mothurOut("The bootstrap.shared command can only be executed after a successful read.otu command.\n"); + m->mothurOut("The bootstrap.shared command parameters are groups, calc, iters and label.\n"); + m->mothurOut("The groups parameter allows you to specify which of the groups in your groupfile you would like included used.\n"); + m->mothurOut("The group names are separated by dashes. The label parameter allows you to select what distance levels you would like trees created for, and is also separated by dashes.\n"); + m->mothurOut("The bootstrap.shared command should be in the following format: bootstrap.shared(groups=yourGroups, calc=yourCalcs, label=yourLabels, iters=yourIters).\n"); + m->mothurOut("Example bootstrap.shared(groups=A-B-C, calc=jabund-sorabund, iters=100).\n"); + m->mothurOut("The default value for groups is all the groups in your groupfile.\n"); + m->mothurOut("The default value for calc is jclass-thetayc. The default for iters is 1000.\n"); + } + catch(exception& e) { + m->errorOut(e, "BootSharedCommand", "help"); exit(1); - } + } } + //********************************************************************************************************************** BootSharedCommand::~BootSharedCommand(){ - delete input; - delete read; - delete util; + //made new in execute + if (abort == false) { + delete input; globaldata->ginput = NULL; + delete read; + delete util; + globaldata->gorder = NULL; + } } //********************************************************************************************************************** int BootSharedCommand::execute(){ try { - int count = 1; - //if the users entered no valid calculators don't execute command - if (treeCalculators.size() == 0) { return 0; } - + if (abort == true) { if (calledHelp) { return 0; } return 2; } + + util = new SharedUtil(); + //read first line read = new ReadOTUFile(globaldata->inputFileName); read->read(&*globaldata); input = globaldata->ginput; order = input->getSharedOrderVector(); - SharedOrderVector* lastOrder = order; + string lastLabel = order->getLabel(); + //if the users entered no valid calculators don't execute command + if (treeCalculators.size() == 0) { return 0; } + //if the users enters label "0.06" and there is no "0.06" in their file use the next lowest label. set processedLabels; - set userLabels = globaldata->labels; + set userLabels = labels; //set users groups util->setGroups(globaldata->Groups, globaldata->gGroupmap->namesOfGroups, "treegroup"); @@ -120,83 +267,111 @@ int BootSharedCommand::execute(){ tmap->makeSim(globaldata->gGroupmap); globaldata->gTreemap = tmap; - while((order != NULL) && ((globaldata->allLines == 1) || (userLabels.size() != 0))) { - - if(globaldata->allLines == 1 || globaldata->lines.count(count) == 1 || globaldata->labels.count(order->getLabel()) == 1){ + while((order != NULL) && ((allLines == 1) || (userLabels.size() != 0))) { + if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } globaldata->Groups.clear(); return 0; } + + if(allLines == 1 || labels.count(order->getLabel()) == 1){ - cout << order->getLabel() << '\t' << count << endl; - process(order); + m->mothurOut(order->getLabel()); m->mothurOutEndLine(); + int error = process(order); + if (error == 1) { for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } globaldata->Groups.clear(); return 0; } processedLabels.insert(order->getLabel()); userLabels.erase(order->getLabel()); - - //you have a label the user want that is smaller than this line and the last line has not already been processed } - if ((anyLabelsToProcess(order->getLabel(), userLabels, "") == true) && (processedLabels.count(lastOrder->getLabel()) != 1)) { - - cout << lastOrder->getLabel() << '\t' << count << endl; - process(lastOrder); + //you have a label the user want that is smaller than this label and the last label has not already been processed + if ((m->anyLabelsToProcess(order->getLabel(), userLabels, "") == true) && (processedLabels.count(lastLabel) != 1)) { + string saveLabel = order->getLabel(); + + delete order; + order = input->getSharedOrderVector(lastLabel); + m->mothurOut(order->getLabel()); m->mothurOutEndLine(); + int error = process(order); + if (error == 1) { for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } globaldata->Groups.clear(); return 0; } - processedLabels.insert(lastOrder->getLabel()); - userLabels.erase(lastOrder->getLabel()); + processedLabels.insert(order->getLabel()); + userLabels.erase(order->getLabel()); + + //restore real lastlabel to save below + order->setLabel(saveLabel); } - if (count != 1) { delete lastOrder; } - lastOrder = order; + + lastLabel = order->getLabel(); //get next line to process + delete order; order = input->getSharedOrderVector(); - count++; } + + if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } globaldata->Groups.clear(); return 0; } + //output error messages about any remaining user labels set::iterator it; bool needToRun = false; for (it = userLabels.begin(); it != userLabels.end(); it++) { - cout << "Your file does not include the label "<< *it; - if (processedLabels.count(lastOrder->getLabel()) != 1) { - cout << ". I will use " << lastOrder->getLabel() << "." << endl; + m->mothurOut("Your file does not include the label " + *it); + if (processedLabels.count(lastLabel) != 1) { + m->mothurOut(". I will use " + lastLabel + "."); m->mothurOutEndLine(); needToRun = true; }else { - cout << ". Please refer to " << lastOrder->getLabel() << "." << endl; + m->mothurOut(". Please refer to " + lastLabel + "."); m->mothurOutEndLine(); } } + if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } globaldata->Groups.clear(); return 0; } + //run last line if you need to if (needToRun == true) { - process(lastOrder); - cout << lastOrder->getLabel() << '\t' << count << endl; + if (order != NULL) { delete order; } + order = input->getSharedOrderVector(lastLabel); + m->mothurOut(order->getLabel()); m->mothurOutEndLine(); + int error = process(order); + if (error == 1) { for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } globaldata->Groups.clear(); return 0; } + + delete order; + } - delete lastOrder; + if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } globaldata->Groups.clear(); return 0; } //reset groups parameter - globaldata->Groups.clear(); globaldata->setGroups(""); + globaldata->Groups.clear(); + + //set first tree file as new current treefile + string currentTree = ""; + itTypes = outputTypes.find("tree"); + if (itTypes != outputTypes.end()) { + if ((itTypes->second).size() != 0) { currentTree = (itTypes->second)[0]; m->setTreeFile(currentTree); } + } + + m->mothurOutEndLine(); + m->mothurOut("Output File Names: "); m->mothurOutEndLine(); + for (int i = 0; i < outputNames.size(); i++) { m->mothurOut(outputNames[i]); m->mothurOutEndLine(); } + m->mothurOutEndLine(); + return 0; } catch(exception& e) { - cout << "Standard Error: " << e.what() << " has occurred in the BootSharedCommand class Function execute. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + m->errorOut(e, "BootSharedCommand", "execute"); exit(1); } - catch(...) { - cout << "An unknown error has occurred in the BootSharedCommand class function execute. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; - exit(1); - } } //********************************************************************************************************************** -void BootSharedCommand::createTree(ostream* out){ +int BootSharedCommand::createTree(ostream* out, Tree* t){ try { - //create tree - t = new Tree(); //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 = -1.0; + if (m->control_pressed) { return 1; } + + 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++) { @@ -229,8 +404,8 @@ void BootSharedCommand::createTree(ostream* out){ index[column] = numGroups+i; //zero out highest value that caused the merge. - simMatrix[row][column] = -1.0; - simMatrix[column][row] = -1.0; + simMatrix[row][column] = -1000.0; + simMatrix[column][row] = -1000.0; //merge values in simsMatrix for (int n = 0; n < simMatrix.size(); n++) { @@ -238,78 +413,83 @@ void BootSharedCommand::createTree(ostream* out){ simMatrix[row][n] = (simMatrix[row][n] + simMatrix[column][n]) / 2; simMatrix[n][row] = simMatrix[row][n]; //delete column - simMatrix[column][n] = -1.0; - simMatrix[n][column] = -1.0; + 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(); + + if (m->control_pressed) { return 1; } //print newick file t->print(*out); - - //delete tree - delete t; + + return 0; } catch(exception& e) { - cout << "Standard Error: " << e.what() << " has occurred in the BootSharedCommand class Function createTree. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; - exit(1); - } - catch(...) { - cout << "An unknown error has occurred in the BootSharedCommand class function createTree. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + m->errorOut(e, "BootSharedCommand", "createTree"); exit(1); } } /***********************************************************/ void BootSharedCommand::printSims() { try { - cout << "simsMatrix" << endl; - for (int m = 0; m < simMatrix.size(); m++) { + m->mothurOut("simsMatrix"); m->mothurOutEndLine(); + for (int k = 0; k < simMatrix.size(); k++) { for (int n = 0; n < simMatrix.size(); n++) { - cout << simMatrix[m][n] << '\t'; + m->mothurOut(toString(simMatrix[k][n])); m->mothurOut("\t"); } - cout << endl; + m->mothurOutEndLine(); } } catch(exception& e) { - cout << "Standard Error: " << e.what() << " has occurred in the BootSharedCommand class Function printSims. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + m->errorOut(e, "BootSharedCommand", "printSims"); exit(1); } - catch(...) { - cout << "An unknown error has occurred in the BootSharedCommand class function printSims. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; - exit(1); - } } /***********************************************************/ -void BootSharedCommand::process(SharedOrderVector* order) { +int BootSharedCommand::process(SharedOrderVector* order) { try{ EstOutput data; vector subset; - + //open an ostream for each calc to print to for (int z = 0; z < treeCalculators.size(); z++) { //create a new filename - outputFile = getRootName(globaldata->inputFileName) + treeCalculators[z]->getName() + ".boot" + order->getLabel() + ".tre"; - openOutputFile(outputFile, *(out[z])); + outputFile = outputDir + m->getRootName(m->getSimpleName(globaldata->inputFileName)) + treeCalculators[z]->getName() + ".boot" + order->getLabel() + ".tre"; + m->openOutputFile(outputFile, *(out[z])); + outputNames.push_back(outputFile); outputTypes["tree"].push_back(outputFile); } + m->mothurOut("Generating bootstrap trees..."); cout.flush(); + //create a file for each calculator with the 1000 trees in it. for (int p = 0; p < iters; p++) { + if (m->control_pressed) { return 1; } + util->getSharedVectorswithReplacement(globaldata->Groups, lookup, order); //fills group vectors from order vector. + //for each calculator for(int i = 0 ; i < treeCalculators.size(); i++) { - + + if (m->control_pressed) { return 1; } + //initialize simMatrix simMatrix.clear(); simMatrix.resize(numGroups); - for (int m = 0; m < simMatrix.size(); m++) { + for (int o = 0; o < simMatrix.size(); o++) { for (int j = 0; j < simMatrix.size(); j++) { - simMatrix[m].push_back(0.0); + simMatrix[o].push_back(0.0); } } @@ -332,23 +512,63 @@ void BootSharedCommand::process(SharedOrderVector* order) { } } } - + + tempTree = new Tree(); + + if (m->control_pressed) { delete tempTree; return 1; } + //creates tree from similarity matrix and write out file - createTree(out[i]); + createTree(out[i], tempTree); + + if (m->control_pressed) { delete tempTree; return 1; } + + //save trees for consensus command. + trees[i].push_back(tempTree); } } + + m->mothurOut("\tDone."); m->mothurOutEndLine(); + //delete globaldata's tree + //for (int m = 0; m < globaldata->gTree.size(); m++) { delete globaldata->gTree[m]; } + //globaldata->gTree.clear(); + + + //create consensus trees for each bootstrapped tree set + for (int k = 0; k < trees.size(); k++) { + + m->mothurOut("Generating consensus tree for " + treeCalculators[k]->getName()); m->mothurOutEndLine(); + + if (m->control_pressed) { return 1; } + + //set global data to calc trees + globaldata->gTree = trees[k]; + + string filename = outputDir + m->getRootName(m->getSimpleName(globaldata->inputFileName)) + treeCalculators[k]->getName() + ".boot" + order->getLabel(); + consensus = new ConcensusCommand(filename); + consensus->execute(); + delete consensus; + + outputNames.push_back(filename + ".cons.pairs"); + outputNames.push_back(filename + ".cons.tre"); + + //delete globaldata's tree + //for (int m = 0; m < globaldata->gTree.size(); m++) { delete globaldata->gTree[m]; } + //globaldata->gTree.clear(); + + } + + + //close ostream for each calc for (int z = 0; z < treeCalculators.size(); z++) { out[z]->close(); } - + + return 0; + } catch(exception& e) { - cout << "Standard Error: " << e.what() << " has occurred in the BootSharedCommand class Function process. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + m->errorOut(e, "BootSharedCommand", "process"); exit(1); } - catch(...) { - cout << "An unknown error has occurred in the BootSharedCommand class function process. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; - exit(1); - } } /***********************************************************/