#include "hclustercommand.h"
//**********************************************************************************************************************
-vector<string> HClusterCommand::getValidParameters(){
+vector<string> HClusterCommand::setParameters(){
try {
- string Array[] = {"cutoff","hard","precision","method","phylip","column","name","sorted","showabund","timing","outputdir","inputdir"};
- vector<string> myArray (Array, Array+(sizeof(Array)/sizeof(string)));
+ CommandParameter pphylip("phylip", "InputTypes", "", "", "PhylipColumn", "PhylipColumn", "none",false,false); parameters.push_back(pphylip);
+ CommandParameter pname("name", "InputTypes", "", "", "none", "none", "ColumnName",false,false); parameters.push_back(pname);
+ CommandParameter pcolumn("column", "InputTypes", "", "", "PhylipColumn", "PhylipColumn", "ColumnName",false,false); parameters.push_back(pcolumn);
+ CommandParameter pcutoff("cutoff", "Number", "", "10", "", "", "",false,false); parameters.push_back(pcutoff);
+ CommandParameter pprecision("precision", "Number", "", "100", "", "", "",false,false); parameters.push_back(pprecision);
+ CommandParameter pmethod("method", "Multiple", "furthest-nearest-average-weighted", "average", "", "", "",false,false); parameters.push_back(pmethod);
+ CommandParameter phard("hard", "Boolean", "", "T", "", "", "",false,false); parameters.push_back(phard);
+ CommandParameter psorted("sorted", "Boolean", "", "F", "", "", "",false,false); parameters.push_back(psorted);
+ CommandParameter pshowabund("showabund", "Boolean", "", "T", "", "", "",false,false); parameters.push_back(pshowabund);
+ CommandParameter ptiming("timing", "Boolean", "", "F", "", "", "",false,false); parameters.push_back(ptiming);
+ CommandParameter pinputdir("inputdir", "String", "", "", "", "", "",false,false); parameters.push_back(pinputdir);
+ CommandParameter poutputdir("outputdir", "String", "", "", "", "", "",false,false); parameters.push_back(poutputdir);
+
+ vector<string> myArray;
+ for (int i = 0; i < parameters.size(); i++) { myArray.push_back(parameters[i].name); }
return myArray;
}
catch(exception& e) {
- m->errorOut(e, "HClusterCommand", "getValidParameters");
+ m->errorOut(e, "HClusterCommand", "setParameters");
exit(1);
}
}
//**********************************************************************************************************************
-HClusterCommand::HClusterCommand(){
+string HClusterCommand::getHelpString(){
try {
- //initialize outputTypes
- vector<string> tempOutNames;
- outputTypes["list"] = tempOutNames;
- outputTypes["rabund"] = tempOutNames;
- outputTypes["sabund"] = tempOutNames;
+ string helpString = "";
+ helpString += "The hcluster command parameter options are cutoff, precision, method, phylip, column, name, showabund, timing and sorted. Phylip or column and name are required, unless you have valid current files.\n";
+ helpString += "The phylip and column parameter allow you to enter your distance file, and sorted indicates whether your column distance file is already sorted. \n";
+ helpString += "The name parameter allows you to enter your name file and is required if your distance file is in column format. \n";
+ helpString += "The hcluster command should be in the following format: \n";
+ helpString += "hcluster(column=youDistanceFile, name=yourNameFile, method=yourMethod, cutoff=yourCutoff, precision=yourPrecision) \n";
+ helpString += "The acceptable hcluster methods are furthest, nearest, weighted and average.\n";
+ return helpString;
}
catch(exception& e) {
- m->errorOut(e, "HClusterCommand", "HClusterCommand");
+ m->errorOut(e, "HClusterCommand", "getHelpString");
exit(1);
}
}
//**********************************************************************************************************************
-vector<string> HClusterCommand::getRequiredParameters(){
+string HClusterCommand::getOutputFileNameTag(string type, string inputName=""){
try {
- string Array[] = {"phylip","column","or"};
- vector<string> myArray (Array, Array+(sizeof(Array)/sizeof(string)));
- return myArray;
+ string outputFileName = "";
+ map<string, vector<string> >::iterator it;
+
+ //is this a type this command creates
+ it = outputTypes.find(type);
+ if (it == outputTypes.end()) { m->mothurOut("[ERROR]: this command doesn't create a " + type + " output file.\n"); }
+ else {
+ if (type == "list") { outputFileName = "list"; }
+ else if (type == "rabund") { outputFileName = "rabund"; }
+ else if (type == "sabund") { outputFileName = "sabund"; }
+ else { m->mothurOut("[ERROR]: No definition for type " + type + " output file tag.\n"); m->control_pressed = true; }
+ }
+ return outputFileName;
}
catch(exception& e) {
- m->errorOut(e, "HClusterCommand", "getRequiredParameters");
+ m->errorOut(e, "HClusterCommand", "getOutputFileNameTag");
exit(1);
}
}
//**********************************************************************************************************************
-vector<string> HClusterCommand::getRequiredFiles(){
+HClusterCommand::HClusterCommand(){
try {
- vector<string> myArray;
- return myArray;
+ abort = true; calledHelp = true;
+ setParameters();
+ vector<string> tempOutNames;
+ outputTypes["list"] = tempOutNames;
+ outputTypes["rabund"] = tempOutNames;
+ outputTypes["sabund"] = tempOutNames;
}
catch(exception& e) {
- m->errorOut(e, "HClusterCommand", "getRequiredFiles");
+ m->errorOut(e, "HClusterCommand", "HClusterCommand");
exit(1);
}
}
//This function checks to make sure the cluster command has no errors and then clusters based on the method chosen.
HClusterCommand::HClusterCommand(string option) {
try{
- globaldata = GlobalData::getInstance();
- abort = false;
+ abort = false; calledHelp = false;
//allow user to run help
- if(option == "help") { help(); abort = true; }
+ if(option == "help") { help(); abort = true; calledHelp = true; }
+ else if(option == "citation") { citation(); abort = true; calledHelp = true;}
else {
- //valid paramters for this command
- string Array[] = {"cutoff","hard","precision","method","phylip","column","name","sorted","showabund","timing","outputdir","inputdir"};
- vector<string> myArray (Array, Array+(sizeof(Array)/sizeof(string)));
+ vector<string> myArray = setParameters();
OptionParser parser(option);
map<string,string> parameters = parser.getParameters();
outputTypes["rabund"] = tempOutNames;
outputTypes["sabund"] = tempOutNames;
- globaldata->newRead();
-
//if the user changes the input directory command factory will send this info to us in the output parameter
string inputDir = validParameter.validFile(parameters, "inputdir", false);
if (inputDir == "not found"){ inputDir = ""; }
phylipfile = validParameter.validFile(parameters, "phylip", true);
if (phylipfile == "not open") { abort = true; }
else if (phylipfile == "not found") { phylipfile = ""; }
- else { distfile = phylipfile; format = "phylip"; }
+ else { distfile = phylipfile; format = "phylip"; m->setPhylipFile(phylipfile); }
columnfile = validParameter.validFile(parameters, "column", true);
if (columnfile == "not open") { abort = true; }
else if (columnfile == "not found") { columnfile = ""; }
- else { distfile = columnfile; format = "column"; }
+ else { distfile = columnfile; format = "column"; m->setColumnFile(columnfile); }
namefile = validParameter.validFile(parameters, "name", true);
if (namefile == "not open") { abort = true; }
else if (namefile == "not found") { namefile = ""; }
+ else { m->setNameFile(namefile); }
- if ((phylipfile == "") && (columnfile == "")) { m->mothurOut("When executing a hcluster command you must enter a phylip or a column."); m->mothurOutEndLine(); abort = true; }
+ if ((phylipfile == "") && (columnfile == "")) {
+ //is there are current file available for either of these?
+ //give priority to column, then phylip
+ columnfile = m->getColumnFile();
+ if (columnfile != "") { m->mothurOut("Using " + columnfile + " as input file for the column parameter."); m->mothurOutEndLine(); }
+ else {
+ phylipfile = m->getPhylipFile();
+ if (phylipfile != "") { m->mothurOut("Using " + phylipfile + " as input file for the phylip parameter."); m->mothurOutEndLine(); }
+ else {
+ m->mothurOut("No valid current files. You must provide a phylip or column file before you can use the hcluster command."); m->mothurOutEndLine();
+ abort = true;
+ }
+ }
+ }
else if ((phylipfile != "") && (columnfile != "")) { m->mothurOut("When executing a hcluster command you must enter ONLY ONE of the following: phylip or column."); m->mothurOutEndLine(); abort = true; }
if (columnfile != "") {
- if (namefile == "") { cout << "You need to provide a namefile if you are going to use the column format." << endl; abort = true; }
+ if (namefile == "") {
+ namefile = m->getNameFile();
+ if (namefile != "") { m->mothurOut("Using " + namefile + " as input file for the name parameter."); m->mothurOutEndLine(); }
+ else {
+ m->mothurOut("You need to provide a namefile if you are going to use the column format."); m->mothurOutEndLine();
+ abort = true;
+ }
+ }
}
//check for optional parameter and set defaults
if (temp == "not found") { temp = "100"; }
//saves precision legnth for formatting below
length = temp.length();
- convert(temp, precision);
+ m->mothurConvert(temp, precision);
- temp = validParameter.validFile(parameters, "hard", false); if (temp == "not found") { temp = "F"; }
+ temp = validParameter.validFile(parameters, "hard", false); if (temp == "not found") { temp = "T"; }
hard = m->isTrue(temp);
temp = validParameter.validFile(parameters, "cutoff", false);
if (temp == "not found") { temp = "10"; }
- convert(temp, cutoff);
+ m->mothurConvert(temp, cutoff);
cutoff += (5 / (precision * 10.0));
method = validParameter.validFile(parameters, "method", false);
- if (method == "not found") { method = "furthest"; }
+ if (method == "not found") { method = "average"; }
- 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 or weighted."); m->mothurOutEndLine(); abort = true; }
showabund = validParameter.validFile(parameters, "showabund", false);
if (showabund == "not found") { showabund = "T"; }
if (method == "furthest") { tag = "fn"; }
else if (method == "nearest") { tag = "nn"; }
+ else if (method == "weighted") { tag = "wn"; }
else { tag = "an"; }
- m->openOutputFile(fileroot+ tag + ".sabund", sabundFile);
- m->openOutputFile(fileroot+ tag + ".rabund", rabundFile);
- m->openOutputFile(fileroot+ tag + ".list", listFile);
-
- 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");
+ string sabundFileName = fileroot+ tag + "." + getOutputFileNameTag("sabund");
+ string rabundFileName = fileroot+ tag + "." + getOutputFileNameTag("rabund");
+ string listFileName = fileroot+ tag + "." + getOutputFileNameTag("list");
+
+ m->openOutputFile(sabundFileName, sabundFile);
+ m->openOutputFile(rabundFileName, rabundFile);
+ m->openOutputFile(listFileName, listFile);
+
+ outputNames.push_back(sabundFileName); outputTypes["sabund"].push_back(sabundFileName);
+ outputNames.push_back(rabundFileName); outputTypes["rabund"].push_back(rabundFileName);
+ outputNames.push_back(listFileName); outputTypes["list"].push_back(listFileName);
}
}
}
//**********************************************************************************************************************
-void HClusterCommand::help(){
- try {
- m->mothurOut("The hcluster command parameter options are cutoff, precision, method, phylip, column, name, showabund, timing and sorted. Phylip or column and name are required.\n");
- m->mothurOut("The phylip and column parameter allow you to enter your distance file, and sorted indicates whether your column distance file is already sorted. \n");
- m->mothurOut("The name parameter allows you to enter your name file and is required if your distance file is in column format. \n");
- m->mothurOut("The hcluster command should be in the following format: \n");
- m->mothurOut("hcluster(column=youDistanceFile, name=yourNameFile, method=yourMethod, cutoff=yourCutoff, precision=yourPrecision) \n");
- m->mothurOut("The acceptable hcluster methods are furthest, nearest and average.\n\n");
- }
- catch(exception& e) {
- m->errorOut(e, "HClusterCommand", "help");
- exit(1);
- }
-}
-
-//**********************************************************************************************************************
-
-HClusterCommand::~HClusterCommand(){}
-
-//**********************************************************************************************************************
-
int HClusterCommand::execute(){
try {
- if (abort == true) { return 0; }
+ if (abort == true) { if (calledHelp) { return 0; } return 2; }
+ NameAssignment* nameMap = NULL;
if(namefile != ""){
- globaldata->nameMap = new NameAssignment(namefile);
- globaldata->nameMap->readMap();
- }else{
- globaldata->nameMap = NULL;
- }
+ nameMap = new NameAssignment(namefile);
+ nameMap->readMap();
+ }
time_t estart = time(NULL);
if (!sorted) {
read = new ReadCluster(distfile, cutoff, outputDir, true);
read->setFormat(format);
- read->read(globaldata->nameMap);
+ read->read(nameMap);
if (m->control_pressed) {
delete read;
sabundFile.close();
rabundFile.close();
listFile.close();
- for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } outputTypes.clear();
+ for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } outputTypes.clear();
return 0;
}
list = read->getListVector();
delete read;
}else {
- list = new ListVector(globaldata->nameMap->getListVector());
+ list = new ListVector(nameMap->getListVector());
}
if (m->control_pressed) {
sabundFile.close();
rabundFile.close();
listFile.close();
- for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } outputTypes.clear();
+ for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } outputTypes.clear();
return 0;
}
print_start = true;
start = time(NULL);
- cluster = new HCluster(rabund, list, method, distfile, globaldata->nameMap, cutoff);
+ cluster = new HCluster(rabund, list, method, distfile, nameMap, cutoff);
vector<seqDist> seqs; seqs.resize(1); // to start loop
if (m->control_pressed) {
sabundFile.close();
rabundFile.close();
listFile.close();
- for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } outputTypes.clear();
+ for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } outputTypes.clear();
return 0;
}
+ float saveCutoff = cutoff;
while (seqs.size() != 0){
seqs = cluster->getSeqs();
+ //to account for cutoff change in average neighbor
+ if (seqs.size() != 0) {
+ if (seqs[0].dist > cutoff) { break; }
+ }
+
if (m->control_pressed) {
delete cluster;
sabundFile.close();
rabundFile.close();
listFile.close();
- for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } outputTypes.clear();
+ for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } outputTypes.clear();
return 0;
}
for (int i = 0; i < seqs.size(); i++) { //-1 means skip me
if (seqs[i].seq1 != seqs[i].seq2) {
- cluster->update(seqs[i].seq1, seqs[i].seq2, seqs[i].dist);
+ cutoff = cluster->update(seqs[i].seq1, seqs[i].seq2, seqs[i].dist);
if (m->control_pressed) {
delete cluster;
sabundFile.close();
rabundFile.close();
listFile.close();
- for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } outputTypes.clear();
+ for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } outputTypes.clear();
return 0;
}
sabundFile.close();
rabundFile.close();
listFile.close();
- for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } outputTypes.clear();
+ for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } outputTypes.clear();
return 0;
}
else if(rndPreviousDist<cutoff){
printData(toString(rndPreviousDist, length-1));
}
-
- //delete globaldata's copy of the sparsematrix and listvector to free up memory
- 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(""); }
-
- globaldata->setListFile(fileroot+ tag + ".list");
- globaldata->setNameFile("");
- globaldata->setFormat("list");
-
+
sabundFile.close();
rabundFile.close();
listFile.close();
delete cluster;
if (m->control_pressed) {
- for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } outputTypes.clear();
+ for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } outputTypes.clear();
return 0;
}
-
+
+
+ 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();
+ }
+
+ //set list file as new current listfile
+ string current = "";
+ itTypes = outputTypes.find("list");
+ if (itTypes != outputTypes.end()) {
+ if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setListFile(current); }
+ }
+
+ //set rabund file as new current rabundfile
+ itTypes = outputTypes.find("rabund");
+ if (itTypes != outputTypes.end()) {
+ if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setRabundFile(current); }
+ }
+
+ //set sabund file as new current sabundfile
+ itTypes = outputTypes.find("sabund");
+ if (itTypes != outputTypes.end()) {
+ if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setSabundFile(current); }
+ }
+
m->mothurOutEndLine();
m->mothurOut("Output File Names: "); m->mothurOutEndLine();