Integrating AI into Biological Research

As computing technologies evolve and advance, so too must the ways we perform scientific research. In a recently released report from the 2025 Workshop on Envisioning Frontiers in AI and Computing for Biological Research, researchers detailed how new technologies such as AI and exascale computing can be used to enhance research in the biological sciences. Pacific Northwest National Laboratory (PNNL) scientists Kirsten Hofmockel and Neeraj Kumar served on the organizing committee for this workshop. 

“One of the big challenges I routinely run into collaborating across domains of science is integrating diverse data across multiple scales to establish genotype to phenotype relationships,” said Hofmockel who leads the Soil Microbiome Science Focus Area project at PNNL. “Individual projects or experiments can widely vary in the amount and diversity of data they produce. Because this data comes in various formats, from images to genetic sequences, it must be integrated in a meaningful way for AI applications.”

The workshop was jointly supported by the Department of Energy’s Advanced Scientific Computing Research (ASCR) program and Biological and Environmental Research (BER) program. During the workshop, participants explored how different techniques, such as multiscale modeling and novel algorithms, can be applied to biological research. They also provided their input on specific areas of research that could benefit most from these techniques in the near future. The resulting report identifies four priority research directions: multimodal data assembly, multiscale biosystems simulation, AI-enabled drivers for experimental systems, and novel algorithms for genomics. The report highlights how combining BER’s extensive efforts in biological data collection and analysis with ASCR’s leading computational capabilities, including exascale architectures and high-performance computing platforms, is an important path to progress.

Co-chaired by Daniela Ushizima of Lawrence Berkeley National Laboratory and Christopher Henry of Argonne National Laboratory, the workshop featured participants from different career stages across academia, industry, and the national laboratory system. PNNL participants included Arunima Bhattacharjee, Aivett Bilbao, William (Bill) Cannon, and Jason McDermott. 

“AI and advanced computing hold immense promise to unlocking breakthroughs in biological research,” said Kumar. “Through close collaboration between computer scientists and domain scientists, we can co-design systems that can enable the next generation of scientific discovery.”

As a chief data scientist in the Advanced Computing, Mathematics, and Data Division and an advisory board member of the Center for AI @PNNL, Kumar leads AI and machine learning programs that advance PNNL’s role in the Department of Energy’s Genesis Mission, the national effort to accelerate scientific discovery through AI-powered platforms. He is driving integration between the Transformational AI Models Consortium and American Science Cloud to build a unified infrastructure for autonomous discovery across biology, chemistry, and critical materials, bridging computational and domain sciences to deliver multidisciplinary impact on a national scale.

Both Kumar and Hofmockel acknowledged the need for innovation in both computing and biological sciences to establish genotype to phenotype relationships and scale-up biological processes.

“We need to innovate algorithms and leverage AI to integrate and interpret diverse biological data,” said Hofmockel. “New collaborations that incorporate biology, advanced computing, and automation are key to advancing the discovery of biological mechanisms and designing new behaviors that support biotechnology and biomanufacturing.”