Configuration
All configuration via environment variables or a .env file in the project root.
Full Reference
Neo4j (KG construction)
| Variable |
Default |
Used by |
NEO4J_URI |
bolt://localhost:7687 |
ingest, ontology build |
NEO4J_USER |
neo4j |
ingest, ontology build |
NEO4J_PASSWORD |
— |
ingest, ontology build |
NEO4J_DATABASE |
neo4j |
ingest, ontology build |
LLM
| Variable |
Default |
Used by |
LLM_SERVICE |
bedrock |
ingest, ontology extract |
LLM_MODEL_ID |
— |
Override default model |
BEDROCK_MODEL_ID |
us.meta.llama3-1-8b-instruct-v1:0 |
Bedrock default |
BEDROCK_REGION |
us-east-1 |
Bedrock API region |
OLLAMA_API_URL |
http://localhost:11434 |
Ollama base URL |
PubMed / PMC
| Variable |
Default |
Used by |
ENTREZ_EMAIL |
bioingest@example.com |
ingest pubmed, ingest pmc |
ENTREZ_API_KEY |
— |
Optional, 10x rate limit |
AWS (Publish)
| Variable |
Default |
Used by |
AWS_PROFILE |
dsinternal |
publish |
AWS_REGION |
eu-north-1 |
publish |
Scraping
| Variable |
Default |
Used by |
FIRECRAWL_API_KEY |
— |
scrape commands |
Data Directory
| Variable |
Default |
Used by |
BIOINGEST_DATA_DIR |
./data |
All commands |
Example .env
# KG construction
NEO4J_URI=bolt://localhost:7687
NEO4J_USER=neo4j
NEO4J_PASSWORD=secret
NEO4J_DATABASE=olink3
# LLM
LLM_SERVICE=bedrock
BEDROCK_REGION=us-east-1
# PubMed
ENTREZ_EMAIL=your@email.com
ENTREZ_API_KEY=your-key
# AWS
AWS_PROFILE=dsinternal
AWS_REGION=eu-north-1
# Scraping
FIRECRAWL_API_KEY=fc-your-key
CLI Overrides
Most config can be overridden via CLI flags:
bioingest ingest pubmed -q "..." --service local --database mydb
bioingest ontology build --uri bolt://remote:7687 --user admin --password pass
bioingest publish --profile other-profile --region us-west-2