diff --git a/src/call_consensus_clustering.cpp b/src/call_consensus_clustering.cpp
index a4fa2fd5..d3d1fa07 100644
--- a/src/call_consensus_clustering.cpp
+++ b/src/call_consensus_clustering.cpp
@@ -176,16 +176,12 @@ void cluster_consensus(std::vector<variant> variants, std::string clustering_fil
    }
    uint32_t position = variants[i].position;
    if(variants[i].low_prob_flag && means[variants[i].cluster_assigned] != (float)0.97 && variants[i].freq < max_mean){
-      std::cerr << "low prob " << position << " " << variants[i].freq << " " << variants[i].nuc << " " << variants[i].probabilities[variants[i].cluster_assigned] << " " << variants[i].cluster_assigned << std::endl;
-      std::cerr << means[variants[i].cluster_assigned] << std::endl;
+      //std::cerr << "low prob " << position << " " << variants[i].freq << " " << variants[i].nuc << " " << variants[i].probabilities[variants[i].cluster_assigned] << " " << variants[i].cluster_assigned << std::endl;
+      //std::cerr << means[variants[i].cluster_assigned] << std::endl;
       continue;
    }
    if(variants[i].vague_assignment && variants[i].freq < 0.97 && variants[i].freq < max_mean){      
-     std::cerr << "vague assignment " << position << " " << variants[i].freq << " " << variants[i].nuc << std::endl;
-    for(auto xx : variants[i].probabilities){
-      std::cerr << xx << " ";
-    }
-    std::cerr << "\n";
+     //std::cerr << "vague assignment " << position << " " << variants[i].freq << " " << variants[i].nuc << std::endl;
     continue;
    }
    auto it = std::find(major_indexes.begin(), major_indexes.end(), variants[i].cluster_assigned);
diff --git a/src/gmm.cpp b/src/gmm.cpp
index 2adbc3d3..3d23e947 100644
--- a/src/gmm.cpp
+++ b/src/gmm.cpp
@@ -164,12 +164,7 @@ std::vector<std::vector<double>> solve_possible_solutions(std::vector<float> tmp
   std::vector<std::vector<double>> viable_solutions;
   for(auto vec : all_combinations){
     double sum = std::accumulate(vec.begin(), vec.end(), 0.0); 
-    /*for(auto x : vec){
-      std::cerr << x << " ";
-    }
-    std::cerr << "sum " << sum << std::endl;*/
     if(sum < 1+error && sum > 1-error && vec.size() > 1) {
-      //std::cerr << "saved" << std::endl;
       viable_solutions.push_back(vec);
     }
   }
@@ -347,6 +342,7 @@ void assign_variants_simple(std::vector<variant> &variants, std::vector<std::vec
   if(smallest_peak < 0.05){
     noise_resampler(n, index, possible_permutations, 3);
   } 
+ 
   //now we loop every unique position and assign the max prob combo of variants
   for(uint32_t i=0; i < unique_pos.size(); i++){
     std::vector<uint32_t> pos_idxs;
@@ -648,13 +644,20 @@ void parse_internal_variants(std::string filename, std::vector<variant> &variant
   } 
 }
 
-std::vector<variant>  gmm_model(std::string prefix, std::vector<uint32_t> populations_iterate, std::string output_prefix){
+std::vector<variant>  gmm_model(std::string prefix, std::vector<uint32_t> populations_iterate, std::string output_prefix, float dcov_first, float dcov_second){
   int retval = 0;
   float lower_bound = 0.01;
   float upper_bound = 0.99;
   uint32_t depth_cutoff = 20;
   float quality_threshold = 20;
-  uint32_t round_val = 4;  
+  uint32_t round_val = 4; 
+
+  float universal_cluster = 0.97;
+  float noise_cluster = 0.03;
+
+  //TESTLINES HEFTS CODE
+  std::vector<std::string> heft_strings;
+
   std::vector<variant> base_variants;
   std::vector<uint32_t> deletion_positions = find_deletion_positions(prefix, depth_cutoff, lower_bound, upper_bound, round_val);
   std::vector<uint32_t> low_quality_positions = find_low_quality_positions(prefix, depth_cutoff, lower_bound, upper_bound, quality_threshold, round_val);
@@ -703,7 +706,6 @@ std::vector<variant>  gmm_model(std::string prefix, std::vector<uint32_t> popula
     if(((float)n > (float)(useful_var/2)) && (n > 2)) continue; //this is because it's recommended to have 10 points per gaussian
     arma::gmm_diag model;
     arma::mat cov (1, n, arma::fill::zeros);
-    std::cerr << "n " << n << " useful var " << useful_var << std::endl;
     bool status = model.learn(data, n, arma::eucl_dist, arma::random_spread, 15, 10, 0.001, false);
     if(status == false){
       std::cerr << "gmm model failed" << std::endl;
@@ -724,11 +726,10 @@ std::vector<variant>  gmm_model(std::string prefix, std::vector<uint32_t> popula
     arma::mat mean_fill (1, n, arma::fill::zeros);
     for(uint32_t l=0; l < n; l++){
       if(l == min_index){
-        mean_fill.col(l) = 0.03;
+        mean_fill.col(l) = noise_cluster;
       } else if(l == max_index){
-        mean_fill.col(l) = 0.97;
-      } else if(means[l] > 0.97){
-        //mean_fill.col(l) = means[l];
+        mean_fill.col(l) = universal_cluster;
+      } else if(means[l] > universal_cluster || means[l] < noise_cluster){
         continue;
       } else{
         mean_fill.col(l) = means[l];
@@ -746,18 +747,18 @@ std::vector<variant>  gmm_model(std::string prefix, std::vector<uint32_t> popula
     for(auto x : model.dcovs){
         tmp_dcovs.push_back(x);
     }
-    std::cerr << model.means << std::endl;
     //found 0.0001, 0.001
     //found 0.00005, 0.001
     //cleaner data 0.0005, 0.001
     for(uint32_t l=0; l < n;l++){
-      if(means[l] >= 0.97 || means[l] <= 0.03){
-        cov.col(l) = 0.0005;
-      } else {
-        cov.col(l) = 0.001;
+      if(means[l] >= universal_cluster){
+        cov.col(l) = dcov_first;
+      } else if (means[l] <= noise_cluster) {
+        cov.col(l) = dcov_first;
+      }else {
+        cov.col(l) = dcov_second;
       }
     }
-    std::cerr << model.dcovs << std::endl;
     model.set_dcovs(cov);
 
     //get the probability of each frequency being assigned to each gaussian
@@ -786,19 +787,70 @@ std::vector<variant>  gmm_model(std::string prefix, std::vector<uint32_t> popula
     double prob_sum = 0;
     determine_low_prob_positions(variants);
 
+    std::vector<variant> retraining_set;
+    std::vector<uint32_t> exclude_cluster_indices;
+    uint32_t retrain_size = 0;
+
+    //std::vector<float> n_hefts;
+    uint32_t new_n = 0;
+    for(uint32_t i=0; i < means.size(); i++){
+      if(means[i] == (float)0.97){
+        exclude_cluster_indices.push_back(i);
+        continue;
+      } else if(means[i] == (float)0.03){
+        exclude_cluster_indices.push_back(i);
+        continue;
+      } else if(means[i] == (float)0.0){
+        exclude_cluster_indices.push_back(i);
+      } else {
+        new_n += 1;
+      }
+    }
+    
     for(uint32_t i=0; i < variants.size(); i++){
-        double prob = variants[i].probabilities[variants[i].cluster_assigned];
-        /*if(variants[i].cluster_assigned == 4){
-          std::cerr << variants[i].freq << " " << variants[i].cluster_assigned << " " << prob << std::endl;
-        }*/
-        prob_sum += prob;
+      auto it = std::find(exclude_cluster_indices.begin(), exclude_cluster_indices.end(), variants[i].cluster_assigned);
+
+      if(it == exclude_cluster_indices.end()){
+        //std::cerr << variants[i].cluster_assigned << " " << variants[i].freq << " " << variants[i].position << " " << variants[i].del_flag << std::endl;
+        retraining_set.push_back(variants[i]);
+        retrain_size += 1;
+      }
+      double prob = variants[i].probabilities[variants[i].cluster_assigned];
+      prob_sum += prob;
+    }
+
+    //retrain the model without things from the universal and noise clusters
+    arma::mat data(1, retrain_size, arma::fill::zeros);
+    //(rows, cols) where each columns is a sample  
+    for(uint32_t i = 0; i < retraining_set.size(); i++){  
+      double tmp = static_cast<double>(retraining_set[i].freq);
+      data.col(i) = tmp;
     }
+    arma::gmm_diag model_final;
+    bool status_2 = model_final.learn(data, new_n, arma::eucl_dist, arma::random_spread, 15, 10, 0.001, false);
     
+    std::cerr << "start mean " << model.means << std::endl; 
+    std::cerr << "start heft " << model.hefts << std::endl; 
+
+    std::cerr << "end mean " << model_final.means << std::endl; 
+    std::cerr << "end heft " << model_final.hefts << std::endl; 
+
+    std::string tmp = "[";
+    for(uint32_t l=0; l < model_final.hefts.size(); l++){
+      if(l != 0) tmp += ",";
+      tmp += std::to_string(model_final.hefts[l]);
+    }
+    tmp += "]";
+    heft_strings.push_back(tmp); 
+    means.clear();
+    for(uint32_t i=0; i < model_final.means.size(); i++){
+      means.push_back((double)model_final.means[i]);
+    }   
+
     solutions.push_back(means);
     models.push_back(model);
     means.clear();  
     double aic = (2 * (double)n) - (2 * prob_sum / (double)useful_var);
-    std::cerr << "aic " << aic << " " << prob_sum << "\n" << std::endl;
     all_aic.push_back(aic);
   }
 
@@ -877,5 +929,33 @@ std::vector<variant>  gmm_model(std::string prefix, std::vector<uint32_t> popula
   file << "means\n";
   file << means_string << "\n";
   file.close();
+
+  std::string cluster_output = output_prefix + "_cluster_data.txt";
+  file.open(cluster_output, std::ios::trunc);
+
+  file << "POS\tALLELE\tFREQ\tCLUSTER\tLIKELIHOOD\n";
+  for(uint32_t i=0; i < variants.size(); i++){
+    file << std::to_string(variants[i].position) << "\t"; 
+    file << variants[i].nuc << "\t";
+    file << std::to_string(variants[i].freq) << "\t";
+    
+    file << std::to_string(variants[i].cluster_assigned) << "\t";
+    if(variants[i].cluster_assigned != -1){
+      float tmp = variants[i].probabilities[variants[i].cluster_assigned];
+      file << std::to_string(tmp) << "\n"; 
+    } else {
+      file << "None\n";
+    }
+  }
+  file.close();
+
+  std::string heft_output = output_prefix + "_hefts.txt";
+  file.open(heft_output, std::ios::trunc);
+  file << "HEFTS\n";
+  for(auto x : heft_strings){
+    file << x << "\n";
+  }
+  file.close();
+
   return(variants);
 }
diff --git a/src/gmm.h b/src/gmm.h
index df29086f..6188d5cd 100644
--- a/src/gmm.h
+++ b/src/gmm.h
@@ -27,7 +27,7 @@ struct variant {
 
 };
 void split(const std::string &s, char delim, std::vector<std::string> &elems);
-std::vector<variant> gmm_model(std::string prefix, std::vector<uint32_t> populations_iterate, std::string output_prefix);
+std::vector<variant> gmm_model(std::string prefix, std::vector<uint32_t> populations_iterate, std::string output_prefix, float dcov_first, float dcov_second);
 void parse_internal_variants(std::string filename, std::vector<variant> &variants, uint32_t depth_cutoff, float lower_bound, float upper_bound, std::vector<uint32_t> deletion_positions, std::vector<uint32_t> low_quality_positions, uint32_t round_val);
 std::vector<uint32_t> find_deletion_positions(std::string filename, uint32_t depth_cutoff, float lower_bound, float upper_bound, uint32_t round_val);
 std::vector<uint32_t> find_low_quality_positions(std::string filename, uint32_t depth_cutoff, float lower_bound, float upper_bound, float quality_threshold, uint32_t round_val);
diff --git a/src/ivar.cpp b/src/ivar.cpp
index 9066345a..e70daba4 100755
--- a/src/ivar.cpp
+++ b/src/ivar.cpp
@@ -48,7 +48,9 @@ struct args_t {
   std::string gff;               // -g
   bool keep_for_reanalysis;      // -k
   //contam params
-  std::string variants;         // -s
+  std::string variants;          // -s
+  float dcov_1;                  // -d
+  float dcov_2;                  // -D
 } g_args;
 
 void print_usage() {
@@ -285,7 +287,7 @@ static const char *removereads_opt_str = "i:p:t:b:h?";
 static const char *filtervariants_opt_str = "p:t:f:h?";
 static const char *getmasked_opt_str = "i:b:f:p:h?";
 static const char *trimadapter_opt_str = "1:2:p:a:h?";
-static const char *contam_opt_str = "i:b:f:x:p:m:q:s:d:e:r:kh?";
+static const char *contam_opt_str = "i:b:f:x:p:m:q:s:d:D:e:r:kh?";
 
 std::string get_filename_without_extension(std::string f, std::string ext) {
   if (ext.length() > f.length())  // If extension longer than filename
@@ -354,6 +356,12 @@ int main(int argc, char *argv[]) {
         case 's':
           g_args.variants = optarg;
           break;
+        case 'd':
+          g_args.dcov_1 = std::stof(optarg);
+          break;
+        case 'D':
+          g_args.dcov_2 = std::stof(optarg);
+          break;
         case 'h':
         case '?':
           print_trim_usage();
@@ -364,10 +372,10 @@ int main(int argc, char *argv[]) {
     }
     if (!g_args.variants.empty() && !g_args.prefix.empty()) {
       std::vector<uint32_t> populations_iterate;
-      for(uint32_t i= 2; i <= 7; i++){
+      for(uint32_t i=6; i <= 6; i++){
         populations_iterate.push_back(i);
       }
-      std::vector<variant> variants = gmm_model(g_args.variants, populations_iterate, g_args.prefix);
+      std::vector<variant> variants = gmm_model(g_args.variants, populations_iterate, g_args.prefix, g_args.dcov_1, g_args.dcov_2);
       cluster_consensus(variants, g_args.prefix);
     }
     res = 0; 
diff --git a/src/primer_bed.cpp b/src/primer_bed.cpp
index ac886522..20924dec 100644
--- a/src/primer_bed.cpp
+++ b/src/primer_bed.cpp
@@ -634,6 +634,9 @@ void primer::transform_mutations(std::string ref_path) {
         ref = true;
       }
       int exists = check_position_exists(current_pos, positions);
+      if(current_pos == 27121 && nuc == "A"){
+        std::cerr << "HERE " << qname << std::endl;
+      }
       if (exists != -1) {
         positions[exists].update_alleles(nuc, ccount, quality[j], ref);  
       } else {