Initialization

After file handling (see Extracting Files), the datasets must be initialized and written into a Zarr store.

• CosMx requires extra preprocessing (renaming columns and generating cell identifiers)

• Xenium can be initialized directly

CosMx Integration

first_run = input("Is it the first run (0: False, 1: True): ")

if first_run == '1':
    # Reload CSVs to apply preprocessing
    fovfile_df = pd.read_csv(flat_file_dir_slide + '/' + fovfile)
    fovfile_df = fovfile_df.rename({'FOV': 'fov'}, axis='columns')
    fovfile_df.to_csv(flat_file_dir_slide + '/' + fovfile, index=False)

    polygon_df = pd.read_csv(flat_file_dir_slide + '/' + polygon)
    polygon_df = polygon_df.rename({'cellID': 'cell_ID'}, axis='columns')
    if 'cell' not in polygon_df.keys():
        polygon_df['cell'] = "c_" + polygon_df['fov'].astype(str) + '_' + polygon_df['cell_ID'].astype(str)
    polygon_df.to_csv(flat_file_dir_slide + '/' + polygon, index=False)

    # Load CosMx data (flip_image may be required depending on dataset version)
    sdata = cosmx.cosmx(cosmx_path + slide, flip_image=True)
    sdata.write(zarr_path)

Xenium Integration

first_run = input("Is it the first run (0: False, 1: True): ")

if first_run == '1':
    # Load Xenium data and save as Zarr
    sdata = xenium.xenium(xenium_path)
    sdata.write(zarr_path)

# Reload data for downstream analysis
sdata = sd.read_zarr(zarr_path)
adata = sdata.tables["table"]
print(adata.obs.keys())

# Example: Spatial plot
xy = adata.obsm['spatial']
plt.scatter(xy[:, 0], xy[:, 1], s=0.0001)

# Example: Quality Control metrics
sc.pp.calculate_qc_metrics(adata, percent_top=(10, 20, 50, 150), inplace=True)
cprobes = (adata.obs["control_probe_counts"].sum() / adata.obs["total_counts"].sum() * 100)
cwords = (adata.obs["control_codeword_counts"].sum() / adata.obs["total_counts"].sum() * 100)
print(f"Negative DNA probe count % : {cprobes}")
print(f"Negative decoding count % : {cwords}")