Add Docker Compose infrastructure, fix API for Memgraph, add ista mapping config

- Fix API bugs: missing lodash import, _whereTemplate naming, hardcoded IPs
- Replace Neo4j APOC calls with Memgraph-compatible schema introspection
- Make DB connection env-driven (COMPTOX_AI_DATABASE_URL)
- Add read-only Bolt proxy (Go) for public Memgraph Lab access
- Add Docker Compose stack: Memgraph, bolt-proxy, API, Lab, Caddy
- Add Caddy reverse proxy config with auto-TLS for api/lab subdomains
- Add EC2 setup script and static site deploy script
- Add GitHub Actions workflow for S3/CloudFront site deployment
- Update Python GraphDB to use bolt:// URI scheme for Memgraph
- Add ista YAML mapping config translating all 31 db.py parse calls
- Add TODO.md for remaining preprocessing tasks

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

Author: JDRomano2

.github/workflows/deploy-site.yml

@@ -0,0 +1,62 @@
+name: Deploy Site
+
+on:
+  push:
+    branches: [master]
+    paths:
+      - 'docs/**'
+      - 'web/packages/app/**'
+
+permissions:
+  id-token: write
+  contents: read
+
+jobs:
+  deploy:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+
+      - name: Set up Python
+        uses: actions/setup-python@v5
+        with:
+          python-version: '3.11'
+
+      - name: Set up Node
+        uses: actions/setup-node@v4
+        with:
+          node-version: '20'
+
+      - name: Install Sphinx dependencies
+        run: |
+          pip install -r docs/requirements.txt 2>/dev/null || pip install sphinx sphinx-rtd-theme
+
+      - name: Build Sphinx docs
+        run: make -C docs html
+
+      - name: Build React app
+        working-directory: web/packages/app
+        run: |
+          npm ci
+          npm run build
+
+      - name: Merge React app into docs
+        run: |
+          mkdir -p docs/build/html/browse
+          cp -r web/packages/app/build/* docs/build/html/browse/
+
+      - name: Configure AWS credentials
+        uses: aws-actions/configure-aws-credentials@v4
+        with:
+          role-to-assume: ${{ secrets.AWS_DEPLOY_ROLE_ARN }}
+          aws-region: us-east-1
+
+      - name: Sync to S3
+        run: aws s3 sync docs/build/html s3://${{ secrets.S3_BUCKET }} --delete
+
+      - name: Invalidate CloudFront
+        if: ${{ secrets.CLOUDFRONT_DIST_ID != '' }}
+        run: |
+          aws cloudfront create-invalidation \
+            --distribution-id ${{ secrets.CLOUDFRONT_DIST_ID }} \
+            --paths "/*"

.gitignore

@@ -106,4 +106,6 @@ scripts/pfhxs
 node_modules
 build
 
+web/packages/api/.env
+
 docs/build/html/

CONFIG-default.yaml

@@ -1,19 +1,14 @@
 ---
 # Default configuration for ComptoxAI
-neo4j:
-  username: neo4j
-  password: neo4j
+database:
+  username: ""
+  password: ""
   hostname: localhost
   protocol: bolt
   port: 7687
 
-mysql:
-  user:
-  passwd:
-  host:
-
 data:
-  prefix: D:\data\
-  tempdir: D:\data\comptox_ai\
+  prefix: /data/
+  tempdir: /tmp/comptox_ai/
 
-...
+...

TODO.md

@@ -0,0 +1,22 @@
+# ComptoxAI TODO
+
+## ista mapping preprocessing tasks
+
+### 1. NCBI Gene `dbXrefs` compound field preprocessing
+The `Homo_sapiens.gene_info` file has a `dbXrefs` column containing pipe-delimited key-value pairs (e.g. `MIM:603369|HGNC:HGNC:1789|Ensembl:ENSG00000123080`). The ista YAML format doesn't support compound field splitting natively. Write a preprocessing script (`comptox_ai/build/scripts/preprocess_ncbi_gene.py`) that reads the TSV and adds separate `MIM`, `HGNC`, and `Ensembl` columns. Then uncomment the three xref property lines in `comptox_ai/build/comptox_mapping.yaml` under the Gene entity_type.
+
+### 2. MolecularInitiatingEvent / AdverseOutcome `append_class`
+The original `db.py` adds MolecularInitiatingEvent and AdverseOutcome as additional OWL class assertions on existing KeyEvent individuals (`append_class=True`). The ista YAML `mode: enrich` adds data properties but may not add class assertions. If ista doesn't support this, run these Cypher queries against Memgraph after `owl2memgraph` import:
+
+```cypher
+// Get MIE event IDs from aopdb.event_info WHERE event_type = 'molecular-initiating-event'
+MATCH (n:KeyEvent) WHERE n.xrefAOPWikiKEID IN [<MIE event_ids>]
+SET n:MolecularInitiatingEvent;
+
+// Get AO event IDs from aopdb.event_info WHERE event_type = 'adverse-outcome'
+MATCH (n:KeyEvent) WHERE n.xrefAOPWikiKEID IN [<AO event_ids>]
+SET n:AdverseOutcome;
+```
+
+### 3. AOP-DB `keIncludedInAOP` multi-value field expansion
+The `event_info` table's `AOP_ids` column contains `"; "`-delimited lists of AOP IDs per KeyEvent row (e.g. `"12; 45; 78"`). If ista doesn't auto-expand multi-value fields during relationship mapping, write a preprocessing step or custom SQL view that explodes each row into one `(event_id, AOP_id)` pair per row before the relationship mapping runs.