feat: update notebook code

notebooks/check_nat2_perf.py

@@ -0,0 +1,80 @@
+import marimo
+
+__generated_with = "0.14.16"
+app = marimo.App(width="medium")
+
+
+@app.cell
+def _():
+    import os
+    import json
+    import marimo as mo
+    import pandas as pd
+    return json, mo, os, pd
+
+
+@app.cell
+def _(mo):
+    mo.md(
+        r"""
+    # Analyse the NAT2 Results post-update of the JSON with the latest star allele mapping.
+
+    ### What changes were made to the pgx-engine
+    - Json updated with the latest [NAT2 allele defintions](https://www.pharmvar.org/gene/NAT2?ref=blog.clinpgx.org)
+        - Used these scripts [pharmvar_to_translation_table.py](https://git.serenomica.com/serenomica/pgx-engine/src/branch/master/scripts/pharmvar_to_translation_table.py) >> [translation_to_json.py](https://git.serenomica.com/serenomica/pgx-engine/src/branch/master/scripts/translation_to_json.py) >> new json
+    - Updated the diplotypes_to_phenotypes.tsv to have the corrected new allele mapping
+        - [Mapping source](https://a.storyblok.com/f/70677/x/301f6834b5/nat2_look-up-table-v1-1.xlsx) PharmVar
+        - Done using [update_nat2_d2p_tsv.py](https://git.serenomica.com/serenomica/pgx-engine/src/branch/fix-pharmvar-to-json-scripts/scripts/update_nat2_d2p_tsv.py)
+    - All validation samples run on server with updated PGx engine
+
+    See [PR#3](https://git.serenomica.com/serenomica/pgx-engine/pulls/3) for more details
+    """
+    )
+    return
+
+
+@app.cell
+def _(pd):
+    # import mapping file
+    nat2_mapping = pd.read_excel("data/nat2_look-up-table-v1-1.xlsx", skiprows=1)
+
+    # paths to the PGx output
+    path_new_nat2 = "data/results_nat2_update_validation/"
+    path_old_nat2 = "../reproduce_setup/data/full_validation_data/"
+
+    # sample list
+    with open("../reproduce_setup/data/se_val_samples.txt", "r") as handle:
+        samples = [
+            s.rstrip().split("/")[-1].replace(".vcf", "")
+            for s in handle.readlines()
+        ]
+    print(f"Number of samples: {len(samples)}")
+    return path_new_nat2, path_old_nat2, samples
+
+
+@app.cell
+def _(json, os, path_new_nat2, path_old_nat2, samples):
+    def parse_result(dir, sample, genes):
+        sample_path = os.path.join(dir, f"{sample}_pgx_result", "output.json")
+
+        with open(sample_path, "r") as handle:
+            sample_data = json.load(handle)
+        return {gene: sample_data["called_genotypes"][gene] for gene in genes}
+
+
+    for sample in samples:
+        print(
+            parse_result(path_old_nat2, sample, ["NAT2"]),
+            "\t",
+            parse_result(path_new_nat2, sample, ["NAT2"]),
+        )
+    return
+
+
+@app.cell
+def _():
+    return
+
+
+if __name__ == "__main__":
+    app.run()