Merge pull request #17 from gwaygenomics/add-feature-select

Add feature select step to recipe

1.generate-profiles/3.feature-select.py

@@ -0,0 +1,68 @@
+import os
+import sys
+import pathlib
+import argparse
+import warnings
+import pandas as pd
+
+from pycytominer import feature_select
+
+sys.path.append("../scripts")
+from config_utils import process_config_file
+
+parser = argparse.ArgumentParser()
+parser.add_argument(
+    "--config_file",
+    help="configuration yaml file for the profiling pipeline",
+    default="profiling_config.yaml",
+)
+args = parser.parse_args()
+config_file = args.config_file
+
+config = process_config_file(config_file)
+
+# Extract config arguments
+core_args = config["core"]
+batch = core_args["batch"]
+float_format = core_args["float_format"]
+compression = core_args["compression"]
+
+normalize_args = config["normalize"]
+feature_select_args = config["feature_select"]
+
+singlecell_from_single_file = core_args["output_one_single_cell_file_only"]
+feature_select_operations = feature_select_args["operations"]
+feature_select_levels = feature_select_args["levels"]
+feature_select_drop_samples = feature_select_args["drop_samples"]
+feature_select_features = feature_select_args["features"]
+feature_select_nacutoff = feature_select_args["na_cutoff"]
+feature_select_corr_threshold = feature_select_args["corr_threshold"]
+feature_select_input_files = normalize_args["normalize_output_files"]
+feature_select_output_files = feature_select_args["feature_select_output_files"]
+
+for data_level in feature_select_levels:
+    if data_level == "single_cell":
+        if not singlecell_from_single_file:
+            warnings.warn(
+                "Feature select operation is not enabled for site-specific single cell files. Skipping."
+            )
+            continue
+
+    input_file = feature_select_input_files[data_level]
+    output_file = feature_select_output_files[data_level]
+
+    print(
+        f"Now performing feature selection for {data_level}...with operations: {feature_select_operations}"
+    )
+
+    feature_select(
+        profiles=input_file,
+        features=feature_select_features,
+        samples=feature_select_drop_samples,
+        operation=feature_select_operations,
+        na_cutoff=feature_select_nacutoff,
+        corr_threshold=feature_select_corr_threshold,
+        output_file=output_file,
+        compression=compression,
+        float_format=float_format,
+    )

1.generate-profiles/profiling_config.yaml

@@ -76,3 +76,20 @@ normalize:
     - single_cell
   by_samples: all
   features: infer
+---
+feature_select:
+  perform: true
+  output_basedir: data/profiles
+  operations:
+    - variance_threshold
+    - drop_na_columns
+    - blacklist
+    - drop_outliers
+  levels:
+    - gene
+    - guide
+    - single_cell
+  drop_samples: none
+  features: infer
+  na_cutoff: 0
+  corr_threshold: 0.9

scripts/config_utils.py

@@ -108,6 +108,19 @@ def generate_profiles_config(config):
             f"{batch}_{normalize_level}_normalized.csv.gz",
         )
 
+    # Build paths to normalize yaml document
+    config["feature_select"]["feature_select_output_dir"] = pathlib.Path(
+        config["feature_select"]["output_basedir"], batch
+    )
+
+    # Build feature select output files
+    config["feature_select"]["feature_select_output_files"] = {}
+    for feature_select_level in config["feature_select"]["levels"]:
+        config["feature_select"]["feature_select_output_files"][feature_select_level] = pathlib.Path(
+            config["feature_select"]["feature_select_output_dir"],
+            f"{batch}_{feature_select_level}_normalized_feature_select.csv.gz",
+        )
+
     return config