University of Utah · Mathematics & Biology

Understanding cellular complexity by learning patterns and modeling mechanisms.

We bring cellular measurements into mathematical and machine learning models to explain biological behavior, predict responses, and guide discovery. We connect observations to mechanisms and learned representations, building models that deepen biological understanding.

Scientific illustration of a crowded human cell cytoplasm, with blue cytoskeletal filaments organizing the surrounding molecules — Order within cellular complexity. Illustration by David S. Goodsell, RCSB Protein Data Bank · CC BY 4.0

What we study

Research programs

Building a bridge from cell measurements to models that explain and predict behavior.

01 /

Whole-cell modeling

Combining mechanistic mathematics with modern machine learning to place cellular measurements inside predictive, interpretable models.

02 /

Discovering causal mechanisms

Combining interventions, cellular measurements, and models to identify causal relationships and uncover mechanisms of biological behavior.

03 /

Foundations of intelligent systems

Studying how AI systems generalize, represent the world, and support scientific reasoning and discovery.

Selected work

Publications

Google Scholar

2026TxPert: using multiple knowledge graphs for prediction of transcriptomic perturbation effectsNature Biotechnology 2025Virtual cells: Predict, explain, discoverarXiv preprint 2024Building, benchmarking, and exploring perturbative maps of transcriptional and morphological dataPLoS Computational Biology 2024MolE: a foundation model for molecular graphs using disentangled attentionNature Communications 2024Masked autoencoders for microscopy are scalable learners of cellular biologyCVPR 2006Biophysical model of AMPA receptor trafficking and its regulation during long-term potentiation and depressionJournal of Neuroscience

Principal investigator

Berton Earnshaw

Research Professor · Department of Mathematics & School of Biological Sciences

Berton is a mathematician and computational scientist whose work connects mechanistic models, AI, and large scale biological data. Before returning to the University of Utah, he helped lead machine learning research for drug discovery at Recursion and Valence Labs.

He earned his PhD in mathematics at Utah in 2007, following undergraduate and master's degrees in mathematics at Brigham Young University. His research now brings together early work in biophysical modeling with experience building data driven systems for biomedical discovery.

He is also a Senior Fellow in Utah's College of Science and serves as a scientific advisor to HealthTree Foundation, Valid Therapeutics, and Blackrock NeuroTech.

Curriculum vitae Read the Utah profile AI Utah 100 honoree

Connect

Let's explore what
comes next.

Interested in this work? Explore the research record and professional profiles, or connect through Utah's mathematics and biology departments.

Google Scholar LinkedIn Department of Mathematics School of Biological Sciences