Version history

{
  "cells": [
    {
      "source": "import cellxgene_census\nimport pandas as pd\n\n# Open Census (latest stable release)\nwith cellxgene_census.open_soma() as census:\n    # Pull obs slice for SEA-AD human cortex\n    obs = census['census_data']['homo_sapiens'].obs\n    sea_ad_obs = obs.read(\n        value_filter=\"dataset_id == 'SEA-AD' or collection_name.str.contains('SEA-AD', case=False, na=False)\",\n        column_names=None  # fetch all columns to enumerate schema\n    ).concat().to_pandas()\n\nprint('SEA-AD obs shape:', sea_ad_obs.shape)\nprint('All obs columns:', list(sea_ad_obs.columns))\nprint('Sample disease values:', sea_ad_obs['disease'].unique() if 'disease' in sea_ad_obs.columns else 'N/A')\nprint('Sample disease_ontology_term_id values:', sea_ad_obs['disease_ontology_term_id'].unique() if 'disease_ontology_term_id' in sea_ad_obs.columns else 'N/A')\n\n# Search for Braak-related columns\nbraak_cols = [c for c in sea_ad_obs.columns if 'braak' in c.lower() or 'neuropath' in c.lower() or 'stage' in c.lower()]\nprint('Braak/neuropath candidate columns:', braak_cols)\n\n# Count microglia donors with Braak metadata\nif braak_cols:\n    microglia_mask = sea_ad_obs['cell_type'].str.contains('microgli', case=False, na=False) if 'cell_type' in sea_ad_obs.columns else pd.Series(False, index=sea_ad_obs.index)\n    micro_df = sea_ad_obs[microglia_mask]\n    for col in braak_cols:\n        non_null = micro_df[col].dropna()\n        n_donors = micro_df.loc[non_null.index, 'donor_id'].nunique() if 'donor_id' in micro_df.columns else 'unknown'\n        print(f'Column {col}: {len(non_null)} microglia nuclei, {n_donors} donors with non-null Braak metadata')",
      "cell_id": "c-4363534d",
      "outputs": [],
      "cell_hash": "sha256:cce4cdfeb908eba453325214e999651afb1dbae4a98f4710e6d331d114f80c7f",
      "cell_type": "code",
      "execution_count": null
    }
  ],
  "metadata": {},
  "owner_ref": "persona-andy-hickl",
  "created_by": "persona-andy-hickl"
}