Experts from 20 consortia are collaborating to build the Human Reference Atlas (HRA), which aims to map the 37 trillion cells in the healthy human body. The HRA Knowledge Graph contains over 6 million nodes and 57 million edges, enabling complex data queries through the HuBMAP portal, HRA Organ Gallery, and other tools. This presentation will explore how HuBMAP, SenNet, and GTEx data are integrated into the HRA to support precision health and medicine at scale. Learn more at https://humanatlas.io and https://humanatlas.io/api.

Over the last three years we have been working on a vast array of data and linking it into a graph database. Using techniques including streaming, intermediate SQL databases and bulk import we have built a database which links mouse and human genes and can be used in a wide range of scientific research. This talk will detail how we did that computationally and show how to use the database.

Biomedical knowledge is uniquely complex and structured, requiring distinct reasoning strategies compared to other scientific disciplines. This diversity calls for flexible approaches that accommodate multiple reasoning strategies while leveraging in-domain knowledge. We introduce KGARevion, a knowledge graph (KG) based agent designed to address the complexity of knowledge-intensive medical queries. Upon receiving a query, KGARevion generates relevant triplets using the LLM knowledge base. These triplets are then verified against a grounded KG to filter out erroneous information and ensure that only accurate, relevant data contribute to the final answer.

SAGE is an easy-to-use platform built to facilitate knowledge generation from multiomics and clinical data. It is based on a rich dataset from which a dedicated Knowledge Graph (KG) was built. Additionally, we trained an LLM to communicate with the KG, including a chatbot functionality to enable wide access to data-generated insights, all without the risk of hallucination. Learn how this platform facilitates scientific discovery for disease understanding. 

Batch Genealogy is a core data product at the heart of the Product Development, Manufacturing, and Supply Chain domains in any Biopharmaceutical company. Batch Xplorer is a vital resource to serve end-to-end manufacturing batch genealogy data needs such as product developability, product market compliance, and quality investigations. Leveraging AWS Neptune RDF graph database technology, the solution provides a comprehensive set of functionalities for data ingestion, profiling, transformation, navigation, retrieval, and analysis. The underlying solution architecture implemented was built integrating internal and external batch data in an RDF Graph Database (AWS Neptune) and with a user interface built in AWS Amplify.

Biomolecular condensates regulate key biological processes, and their dysfunction, or condensatopathies, drives disease. These are novel therapeutic targets for drug discovery. Dewpoint’s AI-powered platform uses graph-based target identification, multi-omics data, and deep learning models to optimize condensate-modifying drugs (c-mods). This approach prioritizes c-mods for diseases like colorectal cancer, addressing key dysfunctions. Dewpoint’s platform supports oncology and neurodegeneration programs, developing innovative small-molecule therapies for high unmet needs.

This talk introduces a cutting-edge framework that integrates large language models (LLMs), ontologies, and graph structures to unify disparate datasets for biomedical research. This unified platform enhances the ability to derive advanced insights through natural language queries, removing the need for expertise in native query languages. Positioned as a bridge between foundational graph technologies and generative AI, this framework offers transformative potential for life sciences applications, accelerating discovery and innovation.

The Unified Biomedical Knowledge Graph (UBKG), funded by multiple NIH awards, is transforming biomedical research by integrating vast datasets for advanced analytics and discovery. This talk explores UBKG’s capabilities, including the Petagraph, a large-scale integration framework published in Nature Scientific Data. Attendees will gain insights into UBKG’s applications, API functionalities, and how it supports precision medicine, AI-driven research, and translational informatics.