X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=clustercommand.cpp;h=6043d39be8d559010e156111f41a856b716cda40;hb=191ae1be0679d5cf4eda950b3b1bf26fb7dd503d;hp=5e6a9b45aea94babdb66d7c928f48b323e88d573;hpb=fdc1f6eaf544f695fc1511f24bddd7e6069c33ba;p=mothur.git diff --git a/clustercommand.cpp b/clustercommand.cpp index 5e6a9b4..6043d39 100644 --- a/clustercommand.cpp +++ b/clustercommand.cpp @@ -9,6 +9,56 @@ #include "clustercommand.h" +//********************************************************************************************************************** +vector ClusterCommand::getValidParameters(){ + try { + string AlignArray[] = {"cutoff","precision","method","showabund","timing","hard","outputdir","inputdir"}; + vector myArray (AlignArray, AlignArray+(sizeof(AlignArray)/sizeof(string))); + return myArray; + } + catch(exception& e) { + m->errorOut(e, "ClusterCommand", "getValidParameters"); + exit(1); + } +} +//********************************************************************************************************************** +ClusterCommand::ClusterCommand(){ + try { + abort = true; + //initialize outputTypes + vector tempOutNames; + outputTypes["list"] = tempOutNames; + outputTypes["rabund"] = tempOutNames; + outputTypes["sabund"] = tempOutNames; + } + catch(exception& e) { + m->errorOut(e, "ClusterCommand", "ClusterCommand"); + exit(1); + } +} +//********************************************************************************************************************** +vector ClusterCommand::getRequiredParameters(){ + try { + vector myArray; + return myArray; + } + catch(exception& e) { + m->errorOut(e, "ClusterCommand", "getRequiredParameters"); + exit(1); + } +} +//********************************************************************************************************************** +vector ClusterCommand::getRequiredFiles(){ + try { + string Array[] = {"phylip","column","or"}; + vector myArray (Array, Array+(sizeof(Array)/sizeof(string))); + return myArray; + } + catch(exception& e) { + m->errorOut(e, "ClusterCommand", "getRequiredFiles"); + exit(1); + } +} //********************************************************************************************************************** //This function checks to make sure the cluster command has no errors and then clusters based on the method chosen. ClusterCommand::ClusterCommand(string option) { @@ -22,7 +72,7 @@ ClusterCommand::ClusterCommand(string option) { else { //valid paramters for this command - string Array[] = {"cutoff","precision","method","showabund","timing","outputdir","inputdir"}; + string Array[] = {"cutoff","precision","method","showabund","timing","hard","outputdir","inputdir"}; vector myArray (Array, Array+(sizeof(Array)/sizeof(string))); OptionParser parser(option); @@ -37,6 +87,12 @@ ClusterCommand::ClusterCommand(string option) { } } + //initialize outputTypes + vector tempOutNames; + outputTypes["list"] = tempOutNames; + outputTypes["rabund"] = tempOutNames; + outputTypes["sabund"] = tempOutNames; + //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 = ""; } @@ -56,16 +112,19 @@ ClusterCommand::ClusterCommand(string option) { length = temp.length(); convert(temp, precision); + temp = validParameter.validFile(parameters, "hard", false); if (temp == "not found") { temp = "F"; } + hard = m->isTrue(temp); + temp = validParameter.validFile(parameters, "cutoff", false); if (temp == "not found") { temp = "10"; } convert(temp, cutoff); - cutoff += (5 / (precision * 10.0)); + cutoff += (5 / (precision * 10.0)); method = validParameter.validFile(parameters, "method", false); if (method == "not found") { method = "furthest"; } - if ((method == "furthest") || (method == "nearest") || (method == "average")) { } - else { m->mothurOut("Not a valid clustering method. Valid clustering algorithms are furthest, nearest or average."); m->mothurOutEndLine(); abort = true; } + if ((method == "furthest") || (method == "nearest") || (method == "average") || (method == "weighted")) { } + else { m->mothurOut("Not a valid clustering method. Valid clustering algorithms are furthest, nearest, average, and weighted."); m->mothurOutEndLine(); abort = true; } showabund = validParameter.validFile(parameters, "showabund", false); if (showabund == "not found") { showabund = "T"; } @@ -88,18 +147,19 @@ ClusterCommand::ClusterCommand(string option) { if (method == "furthest") { cluster = new CompleteLinkage(rabund, list, matrix, cutoff, method); } else if(method == "nearest"){ cluster = new SingleLinkage(rabund, list, matrix, cutoff, method); } else if(method == "average"){ cluster = new AverageLinkage(rabund, list, matrix, cutoff, method); } + else if(method == "weighted"){ cluster = new WeightedLinkage(rabund, list, matrix, cutoff, method); } tag = cluster->getTag(); - if (outputDir == "") { outputDir += hasPath(globaldata->inputFileName); } - fileroot = outputDir + getRootName(getSimpleName(globaldata->inputFileName)); + if (outputDir == "") { outputDir += m->hasPath(globaldata->inputFileName); } + fileroot = outputDir + m->getRootName(m->getSimpleName(globaldata->inputFileName)); - openOutputFile(fileroot+ tag + ".sabund", sabundFile); - openOutputFile(fileroot+ tag + ".rabund", rabundFile); - openOutputFile(fileroot+ tag + ".list", listFile); + m->openOutputFile(fileroot+ tag + ".sabund", sabundFile); + m->openOutputFile(fileroot+ tag + ".rabund", rabundFile); + m->openOutputFile(fileroot+ tag + ".list", listFile); - outputNames.push_back(fileroot+ tag + ".sabund"); - outputNames.push_back(fileroot+ tag + ".rabund"); - outputNames.push_back(fileroot+ tag + ".list"); + outputNames.push_back(fileroot+ tag + ".sabund"); outputTypes["sabund"].push_back(fileroot+ tag + ".sabund"); + outputNames.push_back(fileroot+ tag + ".rabund"); outputTypes["rabund"].push_back(fileroot+ tag + ".rabund"); + outputNames.push_back(fileroot+ tag + ".list"); outputTypes["list"].push_back(fileroot+ tag + ".list"); } } } @@ -114,7 +174,7 @@ ClusterCommand::ClusterCommand(string option) { void ClusterCommand::help(){ try { m->mothurOut("The cluster command can only be executed after a successful read.dist command.\n"); - m->mothurOut("The cluster command parameter options are method, cuttoff, precision, showabund and timing. No parameters are required.\n"); + m->mothurOut("The cluster command parameter options are method, cuttoff, hard, precision, showabund and timing. No parameters are required.\n"); m->mothurOut("The cluster command should be in the following format: \n"); m->mothurOut("cluster(method=yourMethod, cutoff=yourCutoff, precision=yourPrecision) \n"); m->mothurOut("The acceptable cluster methods are furthest, nearest and average. If no method is provided then furthest is assumed.\n\n"); @@ -154,20 +214,20 @@ int ClusterCommand::execute(){ double saveCutoff = cutoff; while (matrix->getSmallDist() < cutoff && matrix->getNNodes() > 0){ - + if (m->control_pressed) { //clean up delete globaldata->gSparseMatrix; globaldata->gSparseMatrix = NULL; delete globaldata->gListVector; globaldata->gListVector = NULL; if (globaldata->getFormat() == "phylip") { globaldata->setPhylipFile(""); } else if (globaldata->getFormat() == "column") { globaldata->setColumnFile(""); } sabundFile.close();rabundFile.close();listFile.close(); - for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } + for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } outputTypes.clear(); return 0; } - if (print_start && isTrue(timing)) { + if (print_start && m->isTrue(timing)) { m->mothurOut("Clustering (" + tag + ") dist " + toString(matrix->getSmallDist()) + "/" - + toString(roundDist(matrix->getSmallDist(), precision)) + + toString(m->roundDist(matrix->getSmallDist(), precision)) + "\t(precision: " + toString(precision) + ", Nodes: " + toString(matrix->getNNodes()) + ")"); cout.flush(); print_start = false; @@ -178,7 +238,12 @@ int ClusterCommand::execute(){ cluster->update(cutoff); float dist = matrix->getSmallDist(); - float rndDist = roundDist(dist, precision); + float rndDist; + if (hard) { + rndDist = m->ceilDist(dist, precision); + }else{ + rndDist = m->roundDist(dist, precision); + } if(previousDist <= 0.0000 && dist != previousDist){ printData("unique"); @@ -193,7 +258,7 @@ int ClusterCommand::execute(){ oldList = *list; } - if (print_start && isTrue(timing)) { + if (print_start && m->isTrue(timing)) { m->mothurOut("Clustering (" + tag + ") for distance " + toString(previousDist) + "/" + toString(rndPreviousDist) + "\t(precision: " + toString(precision) + ", Nodes: " + toString(matrix->getNNodes()) + ")"); cout.flush(); @@ -223,7 +288,12 @@ int ClusterCommand::execute(){ rabundFile.close(); listFile.close(); - if (saveCutoff != cutoff) { m->mothurOut("changed cutoff to " + toString(cutoff)); m->mothurOutEndLine(); } + if (saveCutoff != cutoff) { + if (hard) { saveCutoff = m->ceilDist(saveCutoff, precision); } + else { saveCutoff = m->roundDist(saveCutoff, precision); } + + m->mothurOut("changed cutoff to " + toString(cutoff)); m->mothurOutEndLine(); + } m->mothurOutEndLine(); m->mothurOut("Output File Names: "); m->mothurOutEndLine(); @@ -231,7 +301,7 @@ int ClusterCommand::execute(){ m->mothurOutEndLine(); - //if (isTrue(timing)) { + //if (m->isTrue(timing)) { m->mothurOut("It took " + toString(time(NULL) - estart) + " seconds to cluster"); m->mothurOutEndLine(); //} @@ -248,7 +318,7 @@ int ClusterCommand::execute(){ void ClusterCommand::printData(string label){ try { - if (isTrue(timing)) { + if (m->isTrue(timing)) { m->mothurOut("\tTime: " + toString(time(NULL) - start) + "\tsecs for " + toString(oldRAbund.getNumBins()) + "\tclusters. Updates: " + toString(loops)); m->mothurOutEndLine(); } @@ -257,7 +327,7 @@ void ClusterCommand::printData(string label){ start = time(NULL); oldRAbund.setLabel(label); - if (isTrue(showabund)) { + if (m->isTrue(showabund)) { oldRAbund.getSAbundVector().print(cout); } oldRAbund.print(rabundFile);