X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=clustercommand.cpp;h=6df5faad98cd9a0cd1a78644794cecc3eef0df7d;hb=260ae19c36cb11a53ddc5a75b5e507f8dd8b31d6;hp=8208b7efbde6148c4d5c6c0ad0d090335edd6d0c;hpb=72ebb6fa35b45b149812d47c2b1cb8acaca64659;p=mothur.git diff --git a/clustercommand.cpp b/clustercommand.cpp index 8208b7e..6df5faa 100644 --- a/clustercommand.cpp +++ b/clustercommand.cpp @@ -11,9 +11,10 @@ //********************************************************************************************************************** //This function checks to make sure the cluster command has no errors and then clusters based on the method chosen. -ClusterCommand::ClusterCommand(string option){ +ClusterCommand::ClusterCommand(string option) { try{ globaldata = GlobalData::getInstance(); + abort = false; //allow user to run help @@ -21,7 +22,7 @@ ClusterCommand::ClusterCommand(string option){ else { //valid paramters for this command - string Array[] = {"cutoff","precision","method","showabund","timing"}; + string Array[] = {"cutoff","precision","method","showabund","timing","hard","outputdir","inputdir"}; vector myArray (Array, Array+(sizeof(Array)/sizeof(string))); OptionParser parser(option); @@ -36,9 +37,12 @@ ClusterCommand::ClusterCommand(string option){ } } + //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 = ""; } + //error checking to make sure they read a distance file if ((globaldata->gSparseMatrix == NULL) || (globaldata->gListVector == NULL)) { - mothurOut("Before you use the cluster command, you first need to read in a distance matrix."); mothurOutEndLine(); + m->mothurOut("Before you use the cluster command, you first need to read in a distance matrix."); m->mothurOutEndLine(); abort = true; } @@ -52,16 +56,19 @@ 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 { mothurOut("Not a valid clustering method. Valid clustering algorithms are furthest, nearest or average."); 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"; } @@ -81,21 +88,27 @@ ClusterCommand::ClusterCommand(string option){ } //create cluster - if (method == "furthest") { cluster = new CompleteLinkage(rabund, list, matrix); } - else if(method == "nearest"){ cluster = new SingleLinkage(rabund, list, matrix); } - else if(method == "average"){ cluster = new AverageLinkage(rabund, list, matrix); } + 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(); - - fileroot = getRootName(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"); } } } catch(exception& e) { - errorOut(e, "ClusterCommand", "ClusterCommand"); + m->errorOut(e, "ClusterCommand", "ClusterCommand"); exit(1); } } @@ -104,14 +117,14 @@ ClusterCommand::ClusterCommand(string option){ void ClusterCommand::help(){ try { - mothurOut("The cluster command can only be executed after a successful read.dist command.\n"); - mothurOut("The cluster command parameter options are method, cuttoff, precision, showabund and timing. No parameters are required.\n"); - mothurOut("The cluster command should be in the following format: \n"); - mothurOut("cluster(method=yourMethod, cutoff=yourCutoff, precision=yourPrecision) \n"); - mothurOut("The acceptable cluster methods are furthest, nearest and average. If no method is provided then furthest is assumed.\n\n"); + 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, 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"); } catch(exception& e) { - errorOut(e, "ClusterCommand", "help"); + m->errorOut(e, "ClusterCommand", "help"); exit(1); } } @@ -133,7 +146,7 @@ int ClusterCommand::execute(){ if (abort == true) { return 0; } time_t estart = time(NULL); - int ndist = matrix->getNNodes(); + //int ndist = matrix->getNNodes(); float previousDist = 0.00000; float rndPreviousDist = 0.00000; oldRAbund = *rabund; @@ -142,11 +155,23 @@ int ClusterCommand::execute(){ print_start = true; start = time(NULL); loops = 0; + double saveCutoff = cutoff; while (matrix->getSmallDist() < cutoff && matrix->getNNodes() > 0){ - if (print_start && isTrue(timing)) { - mothurOut("Clustering (" + tag + ") dist " + toString(matrix->getSmallDist()) + "/" - + toString(roundDist(matrix->getSmallDist(), precision)) + + 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()); } + return 0; + } + + if (print_start && m->isTrue(timing)) { + m->mothurOut("Clustering (" + tag + ") dist " + toString(matrix->getSmallDist()) + "/" + + toString(m->roundDist(matrix->getSmallDist(), precision)) + "\t(precision: " + toString(precision) + ", Nodes: " + toString(matrix->getNNodes()) + ")"); cout.flush(); print_start = false; @@ -154,9 +179,15 @@ int ClusterCommand::execute(){ loops++; - cluster->update(); + 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"); @@ -171,8 +202,8 @@ int ClusterCommand::execute(){ oldList = *list; } - if (print_start && isTrue(timing)) { - mothurOut("Clustering (" + tag + ") for distance " + toString(previousDist) + "/" + toString(rndPreviousDist) + 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(); print_start = false; @@ -188,7 +219,7 @@ int ClusterCommand::execute(){ //delete globaldata's copy of the sparsematrix and listvector to free up memory delete globaldata->gSparseMatrix; globaldata->gSparseMatrix = NULL; delete globaldata->gListVector; globaldata->gListVector = NULL; - + //saves .list file so you can do the collect, rarefaction and summary commands without doing a read.list if (globaldata->getFormat() == "phylip") { globaldata->setPhylipFile(""); } else if (globaldata->getFormat() == "column") { globaldata->setColumnFile(""); } @@ -200,13 +231,29 @@ int ClusterCommand::execute(){ sabundFile.close(); rabundFile.close(); listFile.close(); - if (isTrue(timing)) { - mothurOut("It took " + toString(time(NULL) - estart) + " seconds to cluster " + toString(ndist) + " distances"); 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(); + for (int i = 0; i < outputNames.size(); i++) { m->mothurOut(outputNames[i]); m->mothurOutEndLine(); } + m->mothurOutEndLine(); + + + //if (m->isTrue(timing)) { + m->mothurOut("It took " + toString(time(NULL) - estart) + " seconds to cluster"); m->mothurOutEndLine(); + //} + + return 0; } catch(exception& e) { - errorOut(e, "ClusterCommand", "execute"); + m->errorOut(e, "ClusterCommand", "execute"); exit(1); } } @@ -215,16 +262,16 @@ int ClusterCommand::execute(){ void ClusterCommand::printData(string label){ try { - if (isTrue(timing)) { - mothurOut("\tTime: " + toString(time(NULL) - start) + "\tsecs for " + toString(oldRAbund.getNumBins()) - + "\tclusters. Updates: " + toString(loops)); mothurOutEndLine(); + if (m->isTrue(timing)) { + m->mothurOut("\tTime: " + toString(time(NULL) - start) + "\tsecs for " + toString(oldRAbund.getNumBins()) + + "\tclusters. Updates: " + toString(loops)); m->mothurOutEndLine(); } print_start = true; loops = 0; start = time(NULL); oldRAbund.setLabel(label); - if (isTrue(showabund)) { + if (m->isTrue(showabund)) { oldRAbund.getSAbundVector().print(cout); } oldRAbund.print(rabundFile); @@ -234,7 +281,7 @@ void ClusterCommand::printData(string label){ oldList.print(listFile); } catch(exception& e) { - errorOut(e, "ClusterCommand", "printData"); + m->errorOut(e, "ClusterCommand", "printData"); exit(1); }