]> git.donarmstrong.com Git - mothur.git/blobdiff - heatmapsimcommand.cpp
added set.dir command and modified commands to redirect input and output, removed...
[mothur.git] / heatmapsimcommand.cpp
index dc3db89a366269c53d92632275963c9b6f1e88bc..cef790633454b7d8136556d6441a684f0e45e3de 100644 (file)
 
 //**********************************************************************************************************************
 
-HeatMapSimCommand::HeatMapSimCommand(){
+HeatMapSimCommand::HeatMapSimCommand(string option){
        try {
                globaldata = GlobalData::getInstance();
-               validCalculator = new ValidCalculators();
-               heatmap = new HeatMapSim();
-                       
-               int i;
-               for (i=0; i<globaldata->Estimators.size(); i++) {
-                       if (validCalculator->isValidCalculator("heat", globaldata->Estimators[i]) == true) { 
-                               if (globaldata->Estimators[i] == "jabund") {    
-                                       heatCalculators.push_back(new JAbund());
-                               }else if (globaldata->Estimators[i] == "sorabund") { 
-                                       heatCalculators.push_back(new SorAbund());
-                               }else if (globaldata->Estimators[i] == "jclass") { 
-                                       heatCalculators.push_back(new Jclass());
-                               }else if (globaldata->Estimators[i] == "sorclass") { 
-                                       heatCalculators.push_back(new SorClass());
-                               }else if (globaldata->Estimators[i] == "jest") { 
-                                       heatCalculators.push_back(new Jest());
-                               }else if (globaldata->Estimators[i] == "sorest") { 
-                                       heatCalculators.push_back(new SorEst());
-                               }else if (globaldata->Estimators[i] == "thetayc") { 
-                                       heatCalculators.push_back(new ThetaYC());
-                               }else if (globaldata->Estimators[i] == "thetan") { 
-                                       heatCalculators.push_back(new ThetaN());
-                               }else if (globaldata->Estimators[i] == "morisitahorn") { 
-                                       heatCalculators.push_back(new MorHorn());
-                               }else if (globaldata->Estimators[i] == "braycurtis") { 
-                                       heatCalculators.push_back(new BrayCurtis());
+               abort = false;
+               allLines = 1;
+               labels.clear();
+               Groups.clear();
+               Estimators.clear();
+               
+               //allow user to run help
+               if(option == "help") { validCalculator = new ValidCalculators(); help(); abort = true; }
+               
+               else {
+                       //valid paramters for this command
+                       string AlignArray[] =  {"groups","label", "calc","phylip","column","name","outputdir","inputdir"};
+                       vector<string> myArray (AlignArray, AlignArray+(sizeof(AlignArray)/sizeof(string)));
+                       
+                       OptionParser parser(option);
+                       map<string,string> parameters = parser.getParameters();
+                       
+                       ValidParameters validParameter;
+                       map<string,string>::iterator it;
+                       
+                       //check to make sure all parameters are valid for command
+                       for (it = parameters.begin(); it != parameters.end(); it++) { 
+                               if (validParameter.isValidParameter(it->first, myArray, it->second) != true) {  abort = true;  }
+                       }
+                       
+                       format = "";
+                       //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 = "";         }
+                       
+                       //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 = "";          }
+                       else {
+                               string path;
+                               it = parameters.find("phylip");
+                               //user has given a template file
+                               if(it != parameters.end()){ 
+                                       path = hasPath(it->second);
+                                       //if the user has not given a path then, add inputdir. else leave path alone.
+                                       if (path == "") {       parameters["phylip"] = inputDir + it->second;           }
+                               }
+                               
+                               it = parameters.find("column");
+                               //user has given a template file
+                               if(it != parameters.end()){ 
+                                       path = hasPath(it->second);
+                                       //if the user has not given a path then, add inputdir. else leave path alone.
+                                       if (path == "") {       parameters["column"] = inputDir + it->second;           }
+                               }
+                               
+                               it = parameters.find("name");
+                               //user has given a template file
+                               if(it != parameters.end()){ 
+                                       path = hasPath(it->second);
+                                       //if the user has not given a path then, add inputdir. else leave path alone.
+                                       if (path == "") {       parameters["name"] = inputDir + it->second;             }
+                               }
+                       }
+
+                       //required parameters
+                       phylipfile = validParameter.validFile(parameters, "phylip", true);
+                       if (phylipfile == "not open") { abort = true; }
+                       else if (phylipfile == "not found") { phylipfile = ""; }        
+                       else {  format = "phylip";      if (outputDir == "") { outputDir += hasPath(phylipfile); }  }
+                       
+                       columnfile = validParameter.validFile(parameters, "column", true);
+                       if (columnfile == "not open") { abort = true; } 
+                       else if (columnfile == "not found") { columnfile = ""; }
+                       else {  format = "column";      if (outputDir == "") { outputDir += hasPath(columnfile); } }
+                       
+                       namefile = validParameter.validFile(parameters, "name", true);
+                       if (namefile == "not open") { abort = true; }   
+                       else if (namefile == "not found") { namefile = ""; }
+                       
+                       
+                       //error checking on files                       
+                       if ((globaldata->getSharedFile() == "") && ((phylipfile == "") && (columnfile == "")))  { mothurOut("You must run the read.otu command or provide a distance file before running the heatmap.sim command."); mothurOutEndLine(); abort = true; }
+                       else if ((phylipfile != "") && (columnfile != "")) { mothurOut("When running the heatmap.sim command with a distance file you may not use both the column and the phylip parameters."); mothurOutEndLine(); abort = true; }
+                       
+                       if (columnfile != "") {
+                               if (namefile == "") {  mothurOut("You need to provide a namefile if you are going to use the column format."); mothurOutEndLine(); abort = true; }
+                       }
+                       
+                       if (format == "") { format = "shared"; }
+                       
+                       //check for optional parameter and set defaults
+                       // ...at some point should added some additional type checking...
+                       if (format == "shared") {
+                               if (outputDir == "") { outputDir += hasPath(globaldata->getSharedFile()); }
+                               
+                               label = validParameter.validFile(parameters, "label", false);                   
+                               if (label == "not found") { label = ""; }
+                               else { 
+                                       if(label != "all") {  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; 
+                               }
+                               
+                               calc = validParameter.validFile(parameters, "calc", false);                     
+                               if (calc == "not found") { calc = "jest-thetayc";  }
+                               else { 
+                                       if (calc == "default")  {  calc = "jest-thetayc";  }
+                               }
+                               splitAtDash(calc, Estimators);
+                               
+                               groups = validParameter.validFile(parameters, "groups", false);                 
+                               if (groups == "not found") { groups = ""; }
+                               else { 
+                                       splitAtDash(groups, Groups);
+                                       globaldata->Groups = Groups;
                                }
                        }
+                       
+                       if (abort == false) {
+                               validCalculator = new ValidCalculators();
+                       
+                               int i;
+                               for (i=0; i<Estimators.size(); i++) {
+                                       if (validCalculator->isValidCalculator("heat", Estimators[i]) == true) { 
+                                               if (Estimators[i] == "jabund") {        
+                                                       heatCalculators.push_back(new JAbund());
+                                               }else if (Estimators[i] == "sorabund") { 
+                                                       heatCalculators.push_back(new SorAbund());
+                                               }else if (Estimators[i] == "jclass") { 
+                                                       heatCalculators.push_back(new Jclass());
+                                               }else if (Estimators[i] == "sorclass") { 
+                                                       heatCalculators.push_back(new SorClass());
+                                               }else if (Estimators[i] == "jest") { 
+                                                       heatCalculators.push_back(new Jest());
+                                               }else if (Estimators[i] == "sorest") { 
+                                                       heatCalculators.push_back(new SorEst());
+                                               }else if (Estimators[i] == "thetayc") { 
+                                                       heatCalculators.push_back(new ThetaYC());
+                                               }else if (Estimators[i] == "thetan") { 
+                                                       heatCalculators.push_back(new ThetaN());
+                                               }else if (Estimators[i] == "morisitahorn") { 
+                                                       heatCalculators.push_back(new MorHorn());
+                                               }else if (Estimators[i] == "braycurtis") { 
+                                                       heatCalculators.push_back(new BrayCurtis());
+                                               }
+                                       }
+                               }
+                               
+                       }
                }
-               
-               //reset calc for next command
-               globaldata->setCalc("");
 
+                               
 
        }
        catch(exception& e) {
-               cout << "Standard Error: " << e.what() << " has occurred in the HeatMapSimCommand class Function HeatMapSimCommand. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
+               errorOut(e, "HeatMapSimCommand", "HeatMapSimCommand");
                exit(1);
        }
-       catch(...) {
-               cout << "An unknown error has occurred in the HeatMapSimCommand class function HeatMapSimCommand. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
-               exit(1);
-       }       
 }
+
 //**********************************************************************************************************************
 
-HeatMapSimCommand::~HeatMapSimCommand(){
-       delete input;
-       delete read;
-       delete heatmap;
+void HeatMapSimCommand::help(){
+       try {
+               mothurOut("The heatmap.sim command can only be executed after a successful read.otu command, or by providing a distance file.\n");
+               mothurOut("The heatmap.sim command parameters are phylip, column, name, groups, calc and label.  No parameters are required.\n");
+               mothurOut("There are two ways to use the heatmap.sim command. The first is with the read.otu command. \n");
+               mothurOut("With the read.otu command you may use the groups, label and calc parameters. \n");
+               mothurOut("The groups parameter allows you to specify which of the groups in your groupfile you would like included in your heatmap.\n");
+               mothurOut("The group names are separated by dashes. The label parameter allows you to select what distance levels you would like a heatmap created for, and is also separated by dashes.\n");
+               mothurOut("The heatmap.sim command should be in the following format: heatmap.sim(groups=yourGroups, calc=yourCalc, label=yourLabels).\n");
+               mothurOut("Example heatmap.sim(groups=A-B-C, calc=jabund).\n");
+               mothurOut("The default value for groups is all the groups in your groupfile, and all labels in your inputfile will be used.\n");
+               validCalculator->printCalc("heat", cout);
+               mothurOut("The default value for calc is jclass-thetayc.\n");
+               mothurOut("The heatmap.sim command outputs a .svg file for each calculator you choose at each label you specify.\n");
+               mothurOut("The second way to use the heatmap.sim command is with a distance file representing the distance bewteen your groups. \n");
+               mothurOut("Using the command this way, the phylip or column parameter are required, and only one may be used.  If you use a column file the name filename is required. \n");
+               mothurOut("The heatmap.sim command should be in the following format: heatmap.sim(phylip=yourDistanceFile).\n");
+               mothurOut("Example heatmap.sim(phylip=amazonGroups.dist).\n");
+               mothurOut("Note: No spaces between parameter labels (i.e. groups), '=' and parameters (i.e.yourGroups).\n\n");
+
+       }
+       catch(exception& e) {
+               errorOut(e, "HeatMapSimCommand", "help");
+               exit(1);
+       }
 }
 
 //**********************************************************************************************************************
 
+HeatMapSimCommand::~HeatMapSimCommand(){}
+
+//**********************************************************************************************************************
+
 int HeatMapSimCommand::execute(){
        try {
-               int count = 1;  
+       
+               if (abort == true)  { return 0; }
                
+               heatmap = new HeatMapSim(outputDir);
+               
+               if (format == "shared") {
+                       runCommandShared();
+               }else if (format == "phylip") {
+                       globaldata->inputFileName = phylipfile;
+                       runCommandDist();
+               }else if (format == "column") {
+                       globaldata->inputFileName = columnfile;
+                       runCommandDist();
+               }
+               
+               delete heatmap;
+               delete validCalculator;
+
+               return 0;
+       }
+       catch(exception& e) {
+               errorOut(e, "HeatMapSimCommand", "execute");
+               exit(1);
+       }
+}
+
+//**********************************************************************************************************************
+int HeatMapSimCommand::runCommandShared() {
+       try {
                //if the users entered no valid calculators don't execute command
-               if (heatCalculators.size() == 0) { cout << "No valid calculators." << endl; return 0; }
+               if (heatCalculators.size() == 0) { mothurOut("No valid calculators."); mothurOutEndLine(); return 0; }
                
                //you have groups
                read = new ReadOTUFile(globaldata->inputFileName);      
@@ -92,81 +260,214 @@ int HeatMapSimCommand::execute(){
                        
                input = globaldata->ginput;
                lookup = input->getSharedRAbundVectors();
-               vector<SharedRAbundVector*> lastLookup = lookup;
+               string lastLabel = lookup[0]->getLabel();
                
-               if (lookup.size() < 2) { cout << "You have not provided enough valid groups.  I cannot run the command." << endl; return 0;}
+               if (lookup.size() < 2) { mothurOut("You have not provided enough valid groups.  I cannot run the command."); mothurOutEndLine(); return 0;}
                                
                //if the users enters label "0.06" and there is no "0.06" in their file use the next lowest label.
                set<string> processedLabels;
-               set<string> userLabels = globaldata->labels;
-               set<int> userLines = globaldata->lines;
-
+               set<string> userLabels = labels;
                
                //as long as you are not at the end of the file or done wih the lines you want
-               while((lookup[0] != NULL) && ((globaldata->allLines == 1) || (userLabels.size() != 0) || (userLines.size() != 0))) {
+               while((lookup[0] != NULL) && ((allLines == 1) || (userLabels.size() != 0))) {
                
-                       if(globaldata->allLines == 1 || globaldata->lines.count(count) == 1 || globaldata->labels.count(lookup[0]->getLabel()) == 1){                   
+                       if(allLines == 1 || labels.count(lookup[0]->getLabel()) == 1){                  
        
-                               cout << lookup[0]->getLabel() << '\t' << count << endl;
+                               mothurOut(lookup[0]->getLabel()); mothurOutEndLine();
                                heatmap->getPic(lookup, heatCalculators);
                                        
                                processedLabels.insert(lookup[0]->getLabel());
                                userLabels.erase(lookup[0]->getLabel());
-                               userLines.erase(count);
                        }
                                
-                       if ((anyLabelsToProcess(lookup[0]->getLabel(), userLabels, "") == true) && (processedLabels.count(lastLookup[0]->getLabel()) != 1)) {
-                               cout << lastLookup[0]->getLabel() << '\t' << count << endl;
-                               heatmap->getPic(lastLookup, heatCalculators);
+                       if ((anyLabelsToProcess(lookup[0]->getLabel(), userLabels, "") == true) && (processedLabels.count(lastLabel) != 1)) {
+                               string saveLabel = lookup[0]->getLabel();
+                       
+                               for (int i = 0; i < lookup.size(); i++) {  delete lookup[i];  } 
+                               lookup = input->getSharedRAbundVectors(lastLabel);                              
+
+                               mothurOut(lookup[0]->getLabel()); mothurOutEndLine();
+                               heatmap->getPic(lookup, heatCalculators);
                                        
-                               processedLabels.insert(lastLookup[0]->getLabel());
-                               userLabels.erase(lastLookup[0]->getLabel());
+                               processedLabels.insert(lookup[0]->getLabel());
+                               userLabels.erase(lookup[0]->getLabel());
+                               
+                               //restore real lastlabel to save below
+                               lookup[0]->setLabel(saveLabel);
                        }
                                
                        //prevent memory leak
-                       if (count != 1) { for (int i = 0; i < lastLookup.size(); i++) {  delete lastLookup[i];  } }
-                       lastLookup = lookup;                    
+                        
+                       lastLabel = lookup[0]->getLabel();                      
 
                        //get next line to process
+                       for (int i = 0; i < lookup.size(); i++) {  delete lookup[i];  } 
                        lookup = input->getSharedRAbundVectors();                               
-                       count++;
                }
                        
                //output error messages about any remaining user labels
                set<string>::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(lastLookup[0]->getLabel()) != 1) {
-                               cout << ". I will use " << lastLookup[0]->getLabel() << "." << endl;
+                       mothurOut("Your file does not include the label " + *it)
+                       if (processedLabels.count(lastLabel) != 1) {
+                               mothurOut(". I will use " + lastLabel + "."); mothurOutEndLine();
                                needToRun = true;
                        }else {
-                               cout << ". Please refer to " << lastLookup[0]->getLabel() << "." << endl;
+                               mothurOut(". Please refer to " + lastLabel + "."); mothurOutEndLine();
                        }
                }
                
-               //run last line if you need to
+               //run last label if you need to
                if (needToRun == true)  {
-                       cout << lastLookup[0]->getLabel() << '\t' << count << endl;
-                       heatmap->getPic(lastLookup, heatCalculators);
+                       for (int i = 0; i < lookup.size(); i++) {  if (lookup[i] != NULL) { delete lookup[i]; } } 
+                       lookup = input->getSharedRAbundVectors(lastLabel);                              
+
+                       mothurOut(lookup[0]->getLabel()); mothurOutEndLine();
+                       heatmap->getPic(lookup, heatCalculators);
+                       for (int i = 0; i < lookup.size(); i++) {  delete lookup[i];  } 
                }
                
-               for (int i = 0; i < lastLookup.size(); i++) {  delete lastLookup[i];  }
                        
                //reset groups parameter
                globaldata->Groups.clear();  
-               globaldata->setGroups("");
                
+               delete input;  globaldata->ginput = NULL;
+               delete read;
+
                return 0;
        }
        catch(exception& e) {
-               cout << "Standard Error: " << e.what() << " has occurred in the HeatMapSimCommand class Function execute. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
+               errorOut(e, "HeatMapSimCommand", "runCommandShared");
                exit(1);
        }
-       catch(...) {
-               cout << "An unknown error has occurred in the HeatMapSimCommand class function execute. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
-               exit(1);
-       }               
 }
+//**********************************************************************************************************************
+int HeatMapSimCommand::runCommandDist() {
+       try {
+       
+               vector< vector<double> > matrix;
+               vector<string> names;
+               ifstream in;
+               
+               //read distance file and create distance vector and names vector
+               if (format == "phylip") {
+                       //read phylip file
+                       openInputFile(phylipfile, in);
+                       
+                       string name;
+                       int numSeqs;
+                       in >> numSeqs >> name; 
+                       
+                       //save name
+                       names.push_back(name);
+               
+                       //resize the matrix and fill with zeros
+                       matrix.resize(numSeqs); 
+                       for(int i = 0; i < numSeqs; i++) {
+                               matrix[i].resize(numSeqs, 0.0);
+                       }
+                                       
+                       //determine if matrix is square or lower triangle
+                       //if it is square read the distances for the first sequence
+                       char d;
+                       bool square;
+                       while((d=in.get()) != EOF){
+                               
+                               //is d a number meaning its square
+                               if(isalnum(d)){ 
+                                       square = true;
+                                       in.putback(d);
+                                       
+                                       for(int i=0;i<numSeqs;i++){
+                                               in >> matrix[0][i];
+                                       }
+                                       break;
+                               }
+                               
+                               //is d a line return meaning its lower triangle
+                               if(d == '\n'){
+                                       square = false;
+                                       break;
+                               }
+                       }
+                       
+                       //read rest of matrix
+                       if (square == true) { 
+                               for(int i=1;i<numSeqs;i++){
+                                       in >> name;             
+                                       names.push_back(name);
+                                       
+                                       for(int j=0;j<numSeqs;j++) {  in >> matrix[i][j];  }
+                                       gobble(in);
+                               }
+                       }else { 
+                               double dist;
+                               for(int i=1;i<numSeqs;i++){
+                                       in >> name;     
+                                       names.push_back(name);  
+                                       
+                                       for(int j=0;j<i;j++){
+                                               in >> dist;
+                                               matrix[i][j] = dist;  matrix[j][i] = dist;
+                                       }
+                                       gobble(in);
+                               }
+                       }
+                       in.close();
+               }else {
+                       //read names file
+                       NameAssignment* nameMap = new NameAssignment(namefile);
+                       nameMap->readMap();
+                       
+                       //put names in order in vector
+                       for (int i = 0; i < nameMap->size(); i++) {
+                               names.push_back(nameMap->get(i));
+                       }
+                       
+                       //resize matrix
+                       matrix.resize(nameMap->size());
+                       for (int i = 0; i < nameMap->size(); i++) {
+                               matrix[i].resize(nameMap->size(), 0.0);
+                       }
+                       
+                       //read column file
+                       string first, second;
+                       double dist;
+                       openInputFile(columnfile, in);
+                       
+                       while (!in.eof()) {
+                               in >> first >> second >> dist; gobble(in);
+                               
+                               map<string, int>::iterator itA = nameMap->find(first);
+                               map<string, int>::iterator itB = nameMap->find(second);
+                               
+                               if(itA == nameMap->end()){  cerr << "AAError: Sequence '" << first << "' was not found in the names file, please correct\n"; exit(1);  }
+                               if(itB == nameMap->end()){  cerr << "ABError: Sequence '" << second << "' was not found in the names file, please correct\n"; exit(1);  }
+                               
+                               //save distance
+                               matrix[itA->second][itB->second] = dist;
+                               matrix[itB->second][itA->second] = dist;
+                       }
+                       in.close();
+                       
+                       delete nameMap;
+               }
+               
 
+               heatmap->getPic(matrix, names); //vector<vector<double>>, vector<string>
+               
+               return 0;
+       }
+       catch(exception& e) {
+               errorOut(e, "HeatMapSimCommand", "runCommandDist");
+               exit(1);
+       }
+}
 //**********************************************************************************************************************
+
+
+
+
+
+