Removed duplicate unnecessary _make_call_pairs function.

chipathlon/generators/peak_call_generator.py

@@ -125,52 +125,6 @@ class PeakCallGenerator(ModuleGenerator):
         additional_inputs = {}
         return (self.get_markers(run), inputs, additional_inputs)
 
-    def _make_call_pairs(self, run, file_list):
-        """
-        :param run: The run currently being processed.
-        :type run: dict
-        :param file_list: A list of bed files and associated metadata.
-        :type file_list: list
-
-        Construct a list of file tuples where the first file is the treatment
-        and the second is a control.  This function is a little gross,
-        but makes our lives easier later.  We need to randomly pair up control
-        and treatment samples that are from the same experiment.  Each run
-        has two different experiments so we keep them separate from each other.
-        If an experiment has the same number of control and treatment samples
-        we pair them up 1-1.  If an experiment has more controls than
-        treatments, we randomly select from the controls for the treatment
-        files.  If an experiment has less controls than treatments, we use
-        all controls, then randomly select duplicates to match to the other
-        treatment files.
-        """
-        experiment_files = []
-        control_files = []
-        peak_data = {}
-        for output_file in file_list:
-            for experiment_id in run["experiments"]:
-                if experiment_id not in peak_data:
-                    peak_data[experiment_id] = {
-                        "control": [],
-                        "experiment": []
-                    }
-                if output_file["prefix"][:len(experiment_id)] == experiment_id:
-                    if output_file["control_sample_ids"]:
-                        peak_data[experiment_id]["control"].append(output_file)
-                    else:
-                        peak_data[experiment_id]["experiment"].append(output_file)
-        for experiment_id in peak_data:
-            experiment_files += peak_data[experiment_id]["experiment"]
-            if len(peak_data[experiment_id]["experiment"]) < len(peak_data[experiment_id]["control"]):
-                control_files += [random.choice(peak_data[experiment_id]["control"]) for i in range(len(peak_data[experiment_id]["experiment"]))]
-            elif len(peak_data[experiment_id]["experiment"]) > len(peak_data[experiment_id]["control"]):
-                control_files += peak_data[experiment_id]["control"]
-                control_files += [random.choice(peak_data[experiment_id]["control"]) for i in
-                    range(len(peak_data[experiment_id]["experiment"]) - len(peak_data[experiment_id]["control"]))]
-            else:
-                control_files += peak_data[experiment_id]["control"]
-        return zip(experiment_files, control_files)
-
     def _make_call_pairs(self, run, result_list):
         """
         :param run: The run currently being processed.