X-Git-Url: https://git.donarmstrong.com/?p=mothur.git;a=blobdiff_plain;f=clustersplitcommand.cpp;h=95693cc420a7e55d5c9a96a102e3d9e55ad0221c;hp=87d26cea75033c911d4db315265123fdd3dd49de;hb=050a3ff02473a3d4c0980964e1a9ebe52e55d6b8;hpb=eb71e28b7b7afd82540f4a8f0bac9429c5b9d713

diff --git a/clustersplitcommand.cpp b/clustersplitcommand.cpp
index 87d26ce..95693cc 100644
--- a/clustersplitcommand.cpp
+++ b/clustersplitcommand.cpp
@@ -61,7 +61,7 @@ string ClusterSplitCommand::getHelpString(){
         helpString += "The cluster parameter allows you to indicate whether you want to run the clustering or just split the distance matrix, default=t";
 		helpString += "The cutoff parameter allow you to set the distance you want to cluster to, default is 0.25. \n";
 		helpString += "The precision parameter allows you specify the precision of the precision of the distances outputted, default=100, meaning 2 decimal places. \n";
-		helpString += "The method allows you to specify what clustering algorythm you want to use, default=average, option furthest, nearest, or average. \n";
+		helpString += "The method allows you to specify what clustering algorithm you want to use, default=average, option furthest, nearest, or average. \n";
 		helpString += "The splitmethod parameter allows you to specify how you want to split your distance file before you cluster, default=distance, options distance, classify or fasta. \n";
 		helpString += "The taxonomy parameter allows you to enter the taxonomy file for your sequences, this is only valid if you are using splitmethod=classify. Be sure your taxonomy file does not include the probability scores. \n";
 		helpString += "The taxlevel parameter allows you to specify the taxonomy level you want to use to split the distance file, default=3, meaning use the first taxon in each list. \n";
@@ -442,8 +442,7 @@ int ClusterSplitCommand::execute(){
         if (m->debug) { m->mothurOut("[DEBUG]: distName.size() = " + toString(distName.size()) + ".\n"); }
                 
 		//output a merged distance file
-		if (splitmethod == "fasta")		{ createMergedDistanceFile(distName); }
-			
+		//if (splitmethod == "fasta")		{ createMergedDistanceFile(distName); }
 				
 		if (m->control_pressed) { return 0; }
 		
@@ -844,9 +843,12 @@ int ClusterSplitCommand::mergeLists(vector<string> listNames, map<float, int> us
 			if (itLabel->first == -1) { thisLabel = "unique"; }
 			else { thisLabel = toString(itLabel->first,  length-1);  } 
 			
-			outList << thisLabel << '\t' << itLabel->second << '\t';
+			//outList << thisLabel << '\t' << itLabel->second << '\t';
             
             RAbundVector* rabund = NULL;
+            ListVector completeList;
+            completeList.setLabel(thisLabel);
+            
             if (countfile == "") {
                 rabund = new RAbundVector();
                 rabund->setLabel(thisLabel);
@@ -855,7 +857,8 @@ int ClusterSplitCommand::mergeLists(vector<string> listNames, map<float, int> us
 			//add in singletons
 			if (listSingle != NULL) {
 				for (int j = 0; j < listSingle->getNumBins(); j++) {
-					outList << listSingle->get(j) << '\t';
+					//outList << listSingle->get(j) << '\t';
+                    completeList.push_back(listSingle->get(j));
 					if (countfile == "") { rabund->push_back(m->getNumNames(listSingle->get(j))); }
 				}
 			}
@@ -872,7 +875,8 @@ int ClusterSplitCommand::mergeLists(vector<string> listNames, map<float, int> us
 				if (list == NULL) { m->mothurOut("Error merging listvectors in file " + listNames[k]); m->mothurOutEndLine();  }	
 				else {		
 					for (int j = 0; j < list->getNumBins(); j++) {
-						outList << list->get(j) << '\t';
+						//outList << list->get(j) << '\t';
+                        completeList.push_back(list->get(j));
 						if (countfile == "") { rabund->push_back(m->getNumNames(list->get(j))); }
 					}
 					delete list;
@@ -885,7 +889,8 @@ int ClusterSplitCommand::mergeLists(vector<string> listNames, map<float, int> us
                 sabund.print(outSabund);
                 rabund->print(outRabund);
             }
-			outList << endl;
+			//outList << endl;
+            completeList.print(outList);
 			
 			if (rabund != NULL) { delete rabund; }
 		}
@@ -947,7 +952,7 @@ vector<string>  ClusterSplitCommand::createProcesses(vector< map<string, string>
             if ((processToAssign-1) == 1) { m->mothurOut(distName[i].begin()->first + "\n"); }
         }
         
-        //not lets reverse the order of ever other process, so we balance big files running with little ones
+        //now lets reverse the order of ever other process, so we balance big files running with little ones
         for (int i = 0; i < processors; i++) {
             //cout << i << endl;
             int remainder = ((i+1) % processors);
@@ -1169,7 +1174,7 @@ string ClusterSplitCommand::clusterClassicFile(string thisDistFile, string thisN
             cluster->readPhylipFile(thisDistFile, nameMap);
 		}else if (countfile != "") {
             ct = new CountTable();
-            ct->readTable(thisNamefile);
+            ct->readTable(thisNamefile, false);
             cluster->readPhylipFile(thisDistFile, ct);
         }
         tag = cluster->getTag();
@@ -1297,7 +1302,7 @@ string ClusterSplitCommand::clusterFile(string thisDistFile, string thisNamefile
             read->read(nameMap);
 		}else if (countfile != "") {
             ct = new CountTable();
-            ct->readTable(thisNamefile);
+            ct->readTable(thisNamefile, false);
             read->read(ct);
         }else { read->read(nameMap); }
 		
@@ -1325,9 +1330,10 @@ string ClusterSplitCommand::clusterFile(string thisDistFile, string thisNamefile
         m->mothurOutEndLine(); m->mothurOut("Clustering " + thisDistFile); m->mothurOutEndLine();
 		
         //create cluster
-        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);	}
+        float adjust = -1.0;
+        if (method == "furthest")	{	cluster = new CompleteLinkage(rabund, list, matrix, cutoff, method, adjust); }
+        else if(method == "nearest"){	cluster = new SingleLinkage(rabund, list, matrix, cutoff, method, adjust); }
+        else if(method == "average"){	cluster = new AverageLinkage(rabund, list, matrix, cutoff, method, adjust);	}
         tag = cluster->getTag();
 		
         if (outputDir == "") { outputDir += m->hasPath(thisDistFile); }