QIAGEN’s Digital Insights bioinformatics division is set to integrate NVIDIA’s accelerated computing and the BioNeMo platform to enhance AI-driven drug discovery.
The move aims to allow pharmaceutical and biotechnology researchers to use AI more effectively by leveraging enhanced bioinformatics tools and curated knowledge bases.
The integration also aims to help them better understand the biology of diseases, identify promising therapeutic targets, and uncover biomarkers to support the development of new medicines.
Qiagen and Nvidia are tackling the challenge by applying graph-based AI, using retrieval and reasoning methods over biomedical knowledge graphs.
This technology allows researchers to explore evidence across biological systems and supports the creation of agentic, multi-step workflows in the drug discovery process.
Qiagen senior vice-president and product portfolio and innovation head Nitin Sood said: “Qiagen Digital Insights has spent more than 25 years building the biomedical knowledge foundation that researchers rely on to interpret complex biology.
“Through this collaboration with Nvidia, we can accelerate the impact of that knowledge by combining it with advanced AI to help customers improve critical steps in drug discovery, from target identification to biomarker research and hypothesis generation.”
The collaboration targets practical applications throughout the drug discovery process, including identification and validation of targets, repurposing of drugs, biomarker discovery, pathway analysis, and hypothesis generation using multi-omics data.
By integrating curated biomedical knowledge, graph-based AI, and accelerated computing, Qiagen seeks to support research teams to transform complex data into well-informed discovery decisions.
Initial pilot programmes will be available to select partners in the pharmaceutical and biotechnology sectors, with a broader rollout planned after validation.
In October 2025, Eli Lilly and Nvidia announced plans to build a new supercomputer to accelerate drug discovery and cut development cycles.
