GrEBI FAQs

Question 1

What is GrEBI and what problem does it solve?

Accepted Answer

GrEBI (Graphs@EBI) is an HPC pipeline that aggregates diverse biomedical knowledge graphs using ontologies and LLM embeddings. It simplifies complex integrative queries across multiple resources, a task difficult with traditional single-resource APIs.

Question 2

How does GrEBI integrate different knowledge graphs?

Accepted Answer

It employs HPC-driven processes, ontologies for semantic alignment, and LLM embeddings for enhanced representation. This allows it to unify data from major biomedical sources like EMBL-EBI, MONARCH Initiative, and ROBOKOP.

Question 3

What are the main outputs provided by GrEBI?

Accepted Answer

GrEBI offers two primary outputs: materialized query results as gzipped CSV files for easy data loading and analysis, and downloadable Neo4j database exports for running custom queries locally or on HPC.

Question 4

Can I use GrEBI data for my own research or applications?

Accepted Answer

Yes, GrEBI is designed to facilitate querying. You can download the Neo4j database exports to run your own Cypher queries or utilize the materialized CSV results, enabling deep analysis across integrated biomedical data.

Question 5

What kind of biomedical data does GrEBI integrate?

Accepted Answer

GrEBI integrates a comprehensive range of biomedical data, including resources from EMBL-EBI (e.g., IMPC, GWAS Catalog), the MONARCH Initiative (e.g., GO, HPOA), and ROBOKOP (e.g., STRING, CTD), among many others, for broad coverage.

GrEBI

GrEBI

主要功能

使用案例

主要功能

使用案例