Add working snakemake rule for preprocessing

Snakefile

@@ -4,3 +4,22 @@ rule setup:
         writing_path="/ceph/margrie/laura/cimaut/",
     output: "setup_output.txt"
     shell: "python calcium_imaging_automation/core/rules/setup.py {input.datasets_path} {input.writing_path} --folder_read_pattern '2*' --file_read_pattern 'rotation_00001.tif' --file_read_pattern '*.bin' > {output}"
+
+import pandas as pd
+
+paths = pd.read_csv("datasets.csv")
+
+rule all:
+    input:
+        expand("preprocess_output_{index}.txt", index=paths["index"])
+
+rule preprocess:
+    input:
+        lambda wildcards: paths.loc[int(wildcards.index), "read_dataset_path"],
+        lambda wildcards: paths.loc[int(wildcards.index), "write_dataset_path"],
+    output:
+        "preprocess_output_{index}.txt"
+    params:
+        index=lambda wildcards: wildcards.index
+    script:
+        "calcium_imaging_automation/core/rules/preprocess.py"

calcium_imaging_automation/core/rules/preprocess.py

@@ -0,0 +1,24 @@
+from pathlib import Path
+
+from derotation.analysis.metrics import stability_of_most_detected_blob
+from derotation.derotate_batch import derotate
+from snakemake.script import snakemake
+
+try:
+    # Input arguments
+    read_dataset_path = Path(snakemake.input[0])
+    write_dataset_path = Path(snakemake.input[1])
+    output = snakemake.output[0]
+
+    # os.system("module load miniconda")
+    # os.system("source activate /nfs/nhome/live/lporta/.conda/envs/cimat")
+    output_path_dataset = write_dataset_path / "ses-0/funcimg/"
+
+    data = derotate(read_dataset_path, output_path_dataset)
+    metric_measured = stability_of_most_detected_blob(data)
+    with open(output, "w") as f:
+        f.write(f"dataset: {read_dataset_path.stem} metric: {metric_measured}")
+except Exception as e:
+    print(e.args)
+    with open(output, "w") as f:
+        f.write(str(e.args))

calcium_imaging_automation/core/rules/setup.py

@@ -2,6 +2,8 @@
 import shutil
 from pathlib import Path
 
+import pandas as pd
+
 from calcium_imaging_automation.core.reader import ReadAquiredData
 from calcium_imaging_automation.core.writer import DatashuttleWrapper
 
@@ -14,6 +16,7 @@ def setup(raw_data_path, folder_read_pattern, file_read_pattern, output_path):
         print("No derivatives folder found")
 
     print(f"Reading data from {raw_data_path}")
+
     reader = ReadAquiredData(
         raw_data_path,
         folder_read_pattern,
@@ -25,10 +28,20 @@ def setup(raw_data_path, folder_read_pattern, file_read_pattern, output_path):
     print(f"Max of tiffs found: {number_of_tiffs}")
 
     writer = DatashuttleWrapper(output_path)
-    print(f"Dataset names: {reader.dataset_names}")
     writer.create_folders(reader.dataset_names, session_number=number_of_tiffs)
     print("Folders created")
 
+    datasets = pd.DataFrame(
+        {
+            "read_dataset_path": reader.datasets_paths,
+            "write_dataset_path": [
+                writer.get_dataset_path(dt.stem)
+                for dt in reader.datasets_paths
+            ],
+        }
+    )
+    datasets.to_csv("datasets.csv", index=True, index_label="index")
+
 
 if __name__ == "__main__":
     parser = argparse.ArgumentParser(