import pandas as pd
import os

catalog_path = "C:/Users/Swamy/OneDrive - Zafrens/Desktop/Z-Screen Deepdive/Shared_Source_Data_June2026/external_reference/catalog/external_reference_dataset_catalog.csv"
metadata_path = "C:/Users/Swamy/OneDrive - Zafrens/Desktop/Z-Screen Deepdive/Shared_Source_Data_June2026/external_reference/scperturb_downloads/scperturb_summary_metadata.csv"

catalog_df = pd.read_csv(catalog_path)
metadata_df = pd.read_csv(metadata_path)

new_rows = []
for _, row in metadata_df.iterrows():
    new_rows.append({
        "dataset_id": "scperturb_" + row["dataset_id"],
        "source_alias": "scperturb_h5ad_downloads",
        "dataset_type": "scPerturb Single-Cell H5AD",
        "primary_metadata_path": "data/external_reference/scperturb_downloads/scperturb_summary_metadata.csv",
        "primary_signature_path": "data/external_reference/scperturb_downloads/" + row["file"],
        "metadata_rows": row["num_perturbations"],
        "signature_rows": row["num_cells"],
        "signature_columns": row["num_genes"],
        "processing_summary": "Raw H5AD downloaded from scPerturb Zenodo. Native AnnData format, enabling direct comparison."
    })

new_rows_df = pd.DataFrame(new_rows)
updated_catalog = pd.concat([catalog_df, new_rows_df], ignore_index=True)

updated_catalog.to_csv(catalog_path, index=False)
print("Successfully appended 10 datasets to external_reference_dataset_catalog.csv")