New AI standard could significantly accelerate biomedical research

A new international platform, BioContextAI, gives AI systems direct access to specialized databases and bioinformatics tools, making biomedical research faster and more accurate. Researchers from Aarhus University play a central role in the project. This is according to the Faculty of Health at Aarhus University on its website.

“After collecting large amounts of data, we need to understand what the results mean, put them in context with the literature and suggest next steps. It’s a time-consuming process,” says Victor Puelles, Professor at Aarhus University.

For example, the platform can automatically generate full reports on unknown genes based on validated research databases.

“Putting together a report like this would normally take me several hours, but now it can run in the background in minutes,” says Malte Kuehl, postdoc and co-developer of BioContextAI.

Humans still have to make decisions

BioContextAI is built on the Model Context Protocol, which allows tools to be installed in AI models as apps.

“Now you can install the equivalent of apps in your AI systems and give them new features as needed,” says Malte Kuehl.

The platform is being developed in collaboration with researchers from Hamburg, Cambridge and Munich and is seen as the start of a larger international research ecosystem.

While the platform can streamline the research process, researchers emphasize that AI cannot replace human expertise.

“Hallucinations can never be completely ruled out in generative AI and therefore important decisions should always be made by human experts,” says Malte Kuehl.