Brian Kuhlman
University of North Carolina at Chapel Hill, NC·Biochemistry
Proteins do their jobs by folding into specific three-dimensional shapes. Understanding and controlling those shapes matters both for basic biology and for building new molecules for research and medicine.
- Fields
Biochemistry
Biophysics
Computer Science
- Topics
protein design
cold unfolding
protein folding
- Grants
- 2 known grants, through May 2029
What they work on
- Test whether a protein core mixing water-repelling and water-attracting amino acids yields a partly unfolded, water-containing cold state, using de novo designed cold unfolding proteins, biophysical measurements and molecular dynamics simulations.
- Map the energy landscapes and pressure-temperature phase diagrams of designed cold unfolding proteins, testing whether cold-state formation becomes less cooperative as the core becomes more water-attracting.
- Develop computational design protocols for beta-sheet cold unfolding proteins, and redesign natural proteins so they unfold under extreme cold or pressure, to validate the new design principles.
- Apply EvoPro, a pipeline that alternates deep learning structure prediction and sequence design, to create new binders against diverse protein targets that could serve as biosensors or therapeutics.
- Design allosteric protein switches, in which one protein's binding changes another's shape, using sequence design models that fit multiple shapes and use negative design to control binding specificity.
- Improve deep learning sequence design by training graph neural networks that represent side-chain atoms explicitly to predict side-chain positions while optimizing the protein sequence.
More about the research
This research approaches that problem from two directions. One uses designed proteins to learn how cold and high pressure reshape protein structure. The other develops deep learning methods to design new proteins that bind chosen targets, switch shape, and assemble into new complexes.
Why it matters
Designed proteins serve as research reagents and medicines. Better computational design could yield protein binders for biosensors and therapeutics, as well as protein switches for understanding or treating disease. Rules for how proteins behave in cold and under pressure could aid the engineering of cold-adapted proteins and the formulation of protein drugs and vaccines.
Known grants
From NIH and NSF research grants. Labs often have other funding we can't see.
Closest labs
Thomas Szyperski
University at Buffalo, NY·Biophysics
Test whether a protein core mixing hydrophobic (water-avoiding) and hydrophilic (water-loving) parts yields a partly cold-unfolded state that includes water, using computationally designed proteins, biophysical measurements and molecular dynamics simulations.
Alberto Perez
University of Florida, FL·Biophysics
Map the binding partners of the ET domain of BET proteins: identify candidates computationally, rank the strongest with machine learning and physics-based methods, and check top leads with NMR experiments.
Ronald Levy
Temple University, PA·Biophysics
Determine the physical basis of epistasis, where one mutation's effect depends on others, in HIV integrase and capsid under different drugs, using molecular dynamics simulations and virus data to build drug-specific fitness landscapes.
The school
University of North Carolina at Chapel Hill: 72 labs on PI Pages, $1.6B a year on research, 523 research doctorates a year.
Reach out
A start for your email. Fill in the brackets.