else {
//valid paramters for this command
- string Array[] = {"cutoff","precision","method"};
+ string Array[] = {"cutoff","precision","method","showabund","timing","outputdir","inputdir"};
vector<string> myArray (Array, Array+(sizeof(Array)/sizeof(string)));
OptionParser parser(option);
//check to make sure all parameters are valid for command
for (map<string,string>::iterator it = parameters.begin(); it != parameters.end(); it++) {
- if (validParameter.isValidParameter(it->first, myArray, it->second) != true) { abort = true; }
+ if (validParameter.isValidParameter(it->first, myArray, it->second) != true) {
+ abort = true;
+ }
}
+ //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)) {
- cout << "Before you use the cluster command, you first need to read in a distance matrix." << endl; abort = true;
+ mothurOut("Before you use the cluster command, you first need to read in a distance matrix."); mothurOutEndLine();
+ abort = true;
}
//check for optional parameter and set defaults
// ...at some point should added some additional type checking...
//get user cutoff and precision or use defaults
string temp;
- temp = validParameter.validFile(parameters, "precision", false); if (temp == "not found") { temp = "100"; }
+ temp = validParameter.validFile(parameters, "precision", false);
+ if (temp == "not found") { temp = "100"; }
//saves precision legnth for formatting below
length = temp.length();
convert(temp, precision);
- temp = validParameter.validFile(parameters, "cutoff", false); if (temp == "not found") { temp = "10"; }
+ temp = validParameter.validFile(parameters, "cutoff", false);
+ if (temp == "not found") { temp = "10"; }
convert(temp, cutoff);
cutoff += (5 / (precision * 10.0));
- method = validParameter.validFile(parameters, "method", false); if (method == "not found") { method = "furthest"; }
-
+ method = validParameter.validFile(parameters, "method", false);
+ if (method == "not found") { method = "furthest"; }
if ((method == "furthest") || (method == "nearest") || (method == "average")) { }
- else {cout << "Not a valid clustering method. Valid clustering algorithms are furthest, nearest or average." << endl; abort = true; }
+ else { mothurOut("Not a valid clustering method. Valid clustering algorithms are furthest, nearest or average."); mothurOutEndLine(); abort = true; }
+
+ showabund = validParameter.validFile(parameters, "showabund", false);
+ if (showabund == "not found") { showabund = "T"; }
+ timing = validParameter.validFile(parameters, "timing", false);
+ if (timing == "not found") { timing = "F"; }
if (abort == false) {
- //get matrix, list and rabund for execute
+
+ //get matrix, list and rabund for execute
if(globaldata->gSparseMatrix != NULL) { matrix = globaldata->gSparseMatrix; }
if(globaldata->gListVector != NULL){
}
//create cluster
- if(method == "furthest") { cluster = new CompleteLinkage(rabund, list, matrix); tag = "fn"; }
- else if(method == "nearest"){ cluster = new SingleLinkage(rabund, list, matrix); tag = "nn"; }
- else if(method == "average"){ cluster = new AverageLinkage(rabund, list, matrix); tag = "an"; }
- else { cout << "error - not recognized method" << endl; abort = true; }
+ if (method == "furthest") { cluster = new CompleteLinkage(rabund, list, matrix, cutoff); }
+ else if(method == "nearest"){ cluster = new SingleLinkage(rabund, list, matrix, cutoff); }
+ else if(method == "average"){ cluster = new AverageLinkage(rabund, list, matrix, cutoff); }
+ tag = cluster->getTag();
- fileroot = getRootName(globaldata->inputFileName);
+ if (outputDir == "") { outputDir += hasPath(globaldata->inputFileName); }
+ fileroot = outputDir + getRootName(getSimpleName(globaldata->inputFileName));
openOutputFile(fileroot+ tag + ".sabund", sabundFile);
openOutputFile(fileroot+ tag + ".rabund", rabundFile);
openOutputFile(fileroot+ tag + ".list", listFile);
-
-
}
-
}
-
}
catch(exception& e) {
- cout << "Standard Error: " << e.what() << " has occurred in the ClusterCommand class Function ClusterCommand. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
- exit(1);
- }
- catch(...) {
- cout << "An unknown error has occurred in the ClusterCommand class function ClusterCommand. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
+ errorOut(e, "ClusterCommand", "ClusterCommand");
exit(1);
}
}
void ClusterCommand::help(){
try {
- cout << "The cluster command can only be executed after a successful read.dist command." << "\n";
- cout << "The cluster command parameter options are method, cuttoff and precision. No parameters are required." << "\n";
- cout << "The cluster command should be in the following format: " << "\n";
- cout << "cluster(method=yourMethod, cutoff=yourCutoff, precision=yourPrecision) " << "\n";
- cout << "The acceptable cluster methods are furthest, nearest and average. If no method is provided then furthest is assumed." << "\n" << "\n";
+ 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");
}
catch(exception& e) {
- cout << "Standard Error: " << e.what() << " has occurred in the ClusterCommand class Function help. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
+ errorOut(e, "ClusterCommand", "help");
exit(1);
}
- catch(...) {
- cout << "An unknown error has occurred in the ClusterCommand class function help. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
- exit(1);
- }
}
//**********************************************************************************************************************
ClusterCommand::~ClusterCommand(){
- delete cluster;
- delete rabund;
+ if (abort == false) {
+ delete cluster;
+ delete rabund;
+ }
}
//**********************************************************************************************************************
if (abort == true) { return 0; }
+ time_t estart = time(NULL);
+ //int ndist = matrix->getNNodes();
float previousDist = 0.00000;
float rndPreviousDist = 0.00000;
oldRAbund = *rabund;
oldList = *list;
+
+ print_start = true;
+ start = time(NULL);
+ loops = 0;
- float x;
- x=0.1;
- toString(x, 2);
-
- while(matrix->getSmallDist() < cutoff && matrix->getNNodes() > 0){
+ while (matrix->getSmallDist() < cutoff && matrix->getNNodes() > 0){
+ if (print_start && isTrue(timing)) {
+ mothurOut("Clustering (" + tag + ") dist " + toString(matrix->getSmallDist()) + "/"
+ + toString(roundDist(matrix->getSmallDist(), precision))
+ + "\t(precision: " + toString(precision) + ", Nodes: " + toString(matrix->getNNodes()) + ")");
+ cout.flush();
+ print_start = false;
+ }
+
+ loops++;
+
cluster->update();
float dist = matrix->getSmallDist();
float rndDist = roundDist(dist, precision);
oldRAbund = *rabund;
oldList = *list;
}
+
+ if (print_start && isTrue(timing)) {
+ mothurOut("Clustering (" + tag + ") for distance " + toString(previousDist) + "/" + toString(rndPreviousDist)
+ + "\t(precision: " + toString(precision) + ", Nodes: " + toString(matrix->getNNodes()) + ")");
+ cout.flush();
+ print_start = false;
+ }
if(previousDist <= 0.0000){
printData("unique");
globaldata->setNameFile("");
globaldata->setFormat("list");
+ sabundFile.close();
+ rabundFile.close();
+ listFile.close();
+ //if (isTrue(timing)) {
+ mothurOut("It took " + toString(time(NULL) - estart) + " seconds to cluster"); mothurOutEndLine();
+ //}
return 0;
}
catch(exception& e) {
- cout << "Standard Error: " << e.what() << " has occurred in the ClusterCommand class Function execute. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
- exit(1);
- }
- catch(...) {
- cout << "An unknown error has occurred in the ClusterCommand class function execute. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
+ errorOut(e, "ClusterCommand", "execute");
exit(1);
}
-
}
//**********************************************************************************************************************
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();
+ }
+ print_start = true;
+ loops = 0;
+ start = time(NULL);
+
oldRAbund.setLabel(label);
- oldRAbund.getSAbundVector().print(cout);
+ if (isTrue(showabund)) {
+ oldRAbund.getSAbundVector().print(cout);
+ }
oldRAbund.print(rabundFile);
oldRAbund.getSAbundVector().print(sabundFile);
oldList.print(listFile);
}
catch(exception& e) {
- cout << "Standard Error: " << e.what() << " has occurred in the ClusterCommand class Function printData. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
- exit(1);
- }
- catch(...) {
- cout << "An unknown error has occurred in the ClusterCommand class function printData. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
+ errorOut(e, "ClusterCommand", "printData");
exit(1);
}
+
}
//**********************************************************************************************************************