fixing compilation errors by rolling back and then adding in depth cutoff parameter again

Author: cmaceves

src/call_consensus_clustering.cpp

@@ -347,9 +347,9 @@ void cluster_consensus(std::vector<variant> variants, std::string clustering_fil
   float solution_error = 0.05;
   double quality_threshold = 20; 
 
-  std::vector<float> error_rate = cluster_error(variants_file, quality_threshold);
-  float freq_lower_bound = error_rate[0];
-  float freq_upper_bound = error_rate[1];
+  double error_rate = cluster_error(variants_file, quality_threshold, depth_cutoff);
+  float freq_lower_bound = 1-error_rate;
+  float freq_upper_bound = error_rate;
 
   //read in the cluster values
   std::vector<float> means = parse_clustering_results(clustering_file);

src/estimate_error.cpp

@@ -39,14 +39,13 @@ std::vector<std::vector<uint32_t>> determine_outlier_points(std::vector<double>
     return(removal_points);
 }
 
-std::vector<float> cluster_error(std::string filename, double quality_threshold){
+double cluster_error(std::string filename, double quality_threshold, uint32_t depth_cutoff){
   /*
     Here we use clustering to determine the value of the noise.
   */
 
   float lower_bound = 0.50;
   float upper_bound = 0.99;
-  uint32_t depth_cutoff = 10;
   uint32_t round_val = 4;
 
   std::vector<uint32_t> deletion_positions = find_deletion_positions(filename, depth_cutoff, lower_bound, upper_bound, round_val);
@@ -77,7 +76,7 @@ std::vector<float> cluster_error(std::string filename, double quality_threshold)
     count_original += 1;
   }
 
-  uint32_t n = 2;
+  uint32_t n = 3;
   gaussian_mixture_model model = retrain_model(n, data_original, variants_original, 2, 0.00001);
   std::vector<double> means = model.means;
 
@@ -114,9 +113,5 @@ std::vector<float> cluster_error(std::string filename, double quality_threshold)
 
   //get the upper edge of the noise cluster
   auto min_it = std::min_element(cleaned_cluster.begin(), cleaned_cluster.end());
-  std::cerr << "min it " << *min_it << std::endl;
-  std::vector<float> tmp;
-  tmp.push_back(1-(*min_it));
-  tmp.push_back(*min_it);
-  return tmp;
+  return *min_it;
 }

src/estimate_error.h

@@ -3,5 +3,5 @@
 #include "./include/armadillo"
 #ifndef estimate_error
 #define estimate_error
-std::vector<float> cluster_error(std::string filename, double quality_threshold);
+double cluster_error(std::string filename, double quality_threshold, uint32_t depth_cutoff);
 #endif

src/gmm.cpp

@@ -959,10 +959,10 @@ std::vector<variant> gmm_model(std::string prefix, std::string output_prefix){
   uint32_t round_val = 4; 
 
   bool development_mode=true;
-  std::vector<float> error_rate = cluster_error(prefix, quality_threshold);
+  double error_rate = cluster_error(prefix, quality_threshold, depth_cutoff);
 
-  float lower_bound = error_rate[0]+0.001;
-  float upper_bound = error_rate[1]-0.001;
+  float lower_bound = 1-error_rate;
+  float upper_bound = error_rate;
   std::cerr << "lower " << lower_bound << " upper " << upper_bound << std::endl;
   std::vector<variant> base_variants;
   std::vector<uint32_t> deletion_positions = find_deletion_positions(prefix, depth_cutoff, lower_bound, upper_bound, round_val);

src/saga.cpp

@@ -5,9 +5,7 @@
 #include <numeric>
 #include <unordered_map>
 #include <unordered_set>
-#include <chrono>
 #include <tuple>
-using namespace std::chrono;
 
 void parse_cigar(const bam1_t* read1, std::vector<uint32_t> &positions, std::vector<std::string> &bases, std::vector<uint32_t> &qualities, uint32_t total_ref_pos, uint8_t min_qual){
   uint32_t total_query_pos=0;