]> git.donarmstrong.com Git - mothur.git/blobdiff - mgclustercommand.cpp
working on mgcluster count file change
[mothur.git] / mgclustercommand.cpp
index 9c45afe95277b5c7106e3f87f64e60e53b9bf321..6ef968c2248de78e304291f3c781a2baa44ad8ea 100644 (file)
@@ -14,6 +14,8 @@ vector<string> MGClusterCommand::setParameters(){
        try {
                CommandParameter pblast("blast", "InputTypes", "", "", "none", "none", "none",false,true); parameters.push_back(pblast);
                CommandParameter pname("name", "InputTypes", "", "", "none", "none", "none",false,false); parameters.push_back(pname);
+        CommandParameter pcount("count", "InputTypes", "", "", "none", "none", "none",false,false); parameters.push_back(pcount);
+        //CommandParameter plarge("large", "Boolean", "", "F", "", "", "",false,false); parameters.push_back(plarge);
                CommandParameter plength("length", "Number", "", "5", "", "", "",false,false); parameters.push_back(plength);
                CommandParameter ppenalty("penalty", "Number", "", "0.10", "", "", "",false,false); parameters.push_back(ppenalty);
                CommandParameter pcutoff("cutoff", "Number", "", "0.70", "", "", "",false,false); parameters.push_back(pcutoff);
@@ -61,6 +63,28 @@ string MGClusterCommand::getHelpString(){
        }
 }
 //**********************************************************************************************************************
+string MGClusterCommand::getOutputFileNameTag(string type, string inputName=""){       
+       try {
+        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, "MGClusterCommand", "getOutputFileNameTag");
+               exit(1);
+       }
+}
+//**********************************************************************************************************************
 MGClusterCommand::MGClusterCommand(){  
        try {
                abort = true; calledHelp = true; 
@@ -142,11 +166,21 @@ MGClusterCommand::MGClusterCommand(string option) {
                        if (namefile == "not open") { abort = true; }   
                        else if (namefile == "not found") { namefile = ""; }
                        else { m->setNameFile(namefile); }
+            
+            countfile = validParameter.validFile(parameters, "count", true);
+                       if (countfile == "not open") { abort = true; }  
+                       else if (countfile == "not found") { countfile = ""; }
+            else { m->setCountTableFile(countfile); }
+            
+            if (countfile != "" && namefile != "") { m->mothurOut("[ERROR]: Cannot have both a name file and count file. Please use one or the other."); m->mothurOutEndLine(); abort = true; }
                        
                        if ((blastfile == "")) { m->mothurOut("When executing a mgcluster command you must provide a blastfile."); m->mothurOutEndLine(); abort = true; }
                        
                        //check for optional parameter and set defaults
                        string temp;
+            temp = validParameter.validFile(parameters, "large", false);                       if (temp == "not found") { temp = "false"; }            
+                       large = m->isTrue(temp); 
+            
                        temp = validParameter.validFile(parameters, "precision", false);                if (temp == "not found") { temp = "100"; }
                        precisionLength = temp.length();
                        m->mothurConvert(temp, precision); 
@@ -177,7 +211,7 @@ MGClusterCommand::MGClusterCommand(string option) {
                        hclusterWanted = m->isTrue(temp); 
                        
                        temp = validParameter.validFile(parameters, "hard", false);                     if (temp == "not found") { temp = "T"; }
-                       hard = m->isTrue(temp);
+                       hard = m->isTrue(temp);            
                }
 
        }
@@ -189,7 +223,6 @@ MGClusterCommand::MGClusterCommand(string option) {
 //**********************************************************************************************************************
 int MGClusterCommand::execute(){
        try {
-               
                if (abort == true) { if (calledHelp) { return 0; }  return 2;   }
                
                //read names file
@@ -202,15 +235,29 @@ int MGClusterCommand::execute(){
                string tag = "";
                time_t start;
                float previousDist = 0.00000;
-               float rndPreviousDist = 0.00000;
-               
+               float rndPreviousDist = 0.00000; 
+        
                //read blastfile - creates sparsematrices for the distances and overlaps as well as a listvector
                //must remember to delete those objects here since readBlast does not
                read = new ReadBlast(blastfile, cutoff, penalty, length, minWanted, hclusterWanted);
                read->read(nameMap);
-               
-               list = new ListVector(nameMap->getListVector());
-               RAbundVector* rabund = new RAbundVector(list->getRAbundVector());
+        
+        list = new ListVector(nameMap->getListVector());
+        RAbundVector* rabund = NULL;
+        
+        if(countfile != "") {
+            //map<string, int> nameMapCounts = m->readNames(namefile);
+            ct = new CountTable();
+            ct->readTable(countfile);
+            rabund = new RAbundVector();
+            createRabund(ct, list, rabund);
+        }else {
+            rabund = new RAbundVector(list->getRAbundVector());
+        }
+        
+                
+               //list = new ListVector(nameMap->getListVector());
+               //rabund = new RAbundVector(list->getRAbundVector());
                
                if (m->control_pressed) { outputTypes.clear(); delete nameMap; delete read; delete list; delete rabund; return 0; }
                
@@ -223,10 +270,13 @@ int MGClusterCommand::execute(){
                else if (method == "nearest")   { tag = "nn";  }
                else                                                    { tag = "an";  }
                
-               //open output files
-               m->openOutputFile(fileroot+ tag + ".list",  listFile);
-               m->openOutputFile(fileroot+ tag + ".rabund",  rabundFile);
-               m->openOutputFile(fileroot+ tag + ".sabund",  sabundFile);
+        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);
                
                if (m->control_pressed) { 
                        delete nameMap; delete read; delete list; delete rabund; 
@@ -478,8 +528,8 @@ int MGClusterCommand::execute(){
                        m->mothurRemove(overlapFile);
                }
                
-               delete list; 
-               delete rabund;
+               delete list;
+               if (!large) {delete rabund;}
                listFile.close();
                sabundFile.close();
                rabundFile.close();
@@ -493,9 +543,9 @@ int MGClusterCommand::execute(){
                
                m->mothurOutEndLine();
                m->mothurOut("Output File Names: "); m->mothurOutEndLine();
-               m->mothurOut(fileroot+ tag + ".list"); m->mothurOutEndLine();   outputNames.push_back(fileroot+ tag + ".list"); outputTypes["list"].push_back(fileroot+ tag + ".list");
-               m->mothurOut(fileroot+ tag + ".rabund"); m->mothurOutEndLine(); outputNames.push_back(fileroot+ tag + ".rabund"); outputTypes["rabund"].push_back(fileroot+ tag + ".rabund");
-               m->mothurOut(fileroot+ tag + ".sabund"); m->mothurOutEndLine(); outputNames.push_back(fileroot+ tag + ".sabund"); outputTypes["sabund"].push_back(fileroot+ tag + ".sabund");
+               m->mothurOut(listFileName); m->mothurOutEndLine();      outputNames.push_back(listFileName); outputTypes["list"].push_back(listFileName);
+               m->mothurOut(rabundFileName); m->mothurOutEndLine();    outputNames.push_back(rabundFileName); outputTypes["rabund"].push_back(rabundFileName);
+               m->mothurOut(sabundFileName); m->mothurOutEndLine();    outputNames.push_back(sabundFileName); outputTypes["sabund"].push_back(sabundFileName);
                m->mothurOutEndLine();
                
                if (saveCutoff != cutoff) { 
@@ -674,7 +724,33 @@ void MGClusterCommand::sortHclusterFiles(string unsortedDist, string unsortedOve
 
 //**********************************************************************************************************************
 
+void MGClusterCommand::createRabund(CountTable*& ct, ListVector*& list, RAbundVector*& rabund){
+    try {
+        //vector<string> names = ct.getNamesOfSeqs();
 
+        //for ( int i; i < ct.getNumGroups(); i++ ) {    rav.push_back( ct.getNumSeqs(names[i]) );    }
+        //return rav;
+        
+        for(int i = 0; i < list->getNumBins(); i++) { 
+           vector<string> binNames;
+           string bin = list->get(i);
+           m->splitAtComma(bin, binNames);
+           int total = 0;
+           for (int j = 0; j < binNames.size(); j++) { 
+               total += ct->getNumSeqs(binNames[j]);
+           }
+           rabund->push_back(total);   
+       }
+        
+        
+    }
+    catch(exception& e) {
+               m->errorOut(e, "MGClusterCommand", "createRabund");
+               exit(1);
+       }
+    
+}
 
+//**********************************************************************************************************************