William F. DeGrado

This study identifies features contributing to catalytic efficiency in a de novo enzyme through computational design and directed evolution, showing how stability and dynamics interplay in enzyme optimization.

This paper demonstrates that large language models can generate de novo protein sequences that bind specific ligands without copying existing sequences, with a 25% hit rate in experimental validation.

This paper designs a de novo protein fold that binds doxorubicin with high affinity through aromatic π-stacking, revealing a new 5-helix globular fold.

This paper introduces EYClamp, a de novo designed enzyme for photo-proximity labeling of E3 ligase neighborhoods in live cells, identifying over 1,500 candidate neighbors for four E3 ligases.

This study shows that random sequences repeated in tandem are more likely to fold into stable structures than fully random sequences, with some structures being novel and experimentally confirmed.

This study shows how a designed generalist binding protein can evolve into specific functions like a fluorophore binder and an efficient enzyme through weak, promiscuous interactions.

The paper introduces NOVARK, an ultrabright biosensor for dynamic imaging of kinase activity, showing it can detect phosphorylation-induced assembly with higher brightness and dynamic range than existing methods.

This study identifies distinct tau filament structures in FTLD-tau caused by the MAPT S305I mutation, revealing a unique 4-layer architecture and potential RNA-like polyanion binding.

An all-atom diffusion model unifying biomolecular structure prediction and sequence-structure generation designs experimentally validated nanobodies, mini-proteins, and linear or cyclic peptides across 26 diverse targets.

This paper designs and evolves de novo helical bundle proteins to catalyze enantioselective Ge-H insertion, showing improved efficiency and selectivity through computational analysis and directed evolution.

This paper designs de novo proteins to modify nanopore complexes, creating customized structural and functional properties for enhanced sequencing applications.

The paper develops a high-throughput method to identify fluoroprobes that distinguish between tau fibril polymorphs using paDSF and chemical screening.

The paper develops a cooperative binding model for ligands to protein fibrils, relating measured EC50 and IC50 values to intrinsic affinity and cooperativity.

This paper designs de novo Rhodamine Binder (Rhobin) tags with high brightness, stability, and applicability in live-cell imaging and extreme conditions.

The paper demonstrates the feasibility of engineering a transmembrane histidine kinase with a de novo metal-binding sensor domain that transmits binding information.

ProBASS predicts mutation effects on PPI binding affinity using sequence and structural features from ESM2 and ESM-IF1, achieving high correlation with experimental data.

This paper designs a protein to bind naphthalenediimides and evaluates its ability to generate reactive species for proximity labeling.

The paper develops a method to determine sequence-dependent propensities for membrane protein insertion states using glycosylation patterns and UMAP analysis.

This paper shows IL-13 and IL-17A activate β1 integrins through an NF-κB/Rho kinase/PIP5K1γ pathway, enhancing force transmission in airway smooth muscle and linking this to asthma pathology.

This study shows Zfp106 binds to G-quadruplex RNAs formed by GGGGCC repeats, inhibiting RAN translation and RNA foci formation in C9orf72-related diseases.

This study identifies an active mechanical mechanism involving PDGFRA+ mesenchymal cells that pattern and fold intestinal villi through myosin II-dependent forces and matrix metalloproteinase-mediated tissue fluidization.

This paper designs drug-binding proteins with low nanomolar affinity and predictable specificity using computational methods, validated by experimental measurements and structural analysis.

SymDOCK is a docking method that exploits symmetry in amyloid fibrils to predict ligand stacking, successfully reproducing known structures and predicting a new one before its determination.

This paper introduces a computational method to design peptides that bind specific conformers of α-synuclein fibrils, resulting in a peptide that slows amyloid formation and shows high specificity.

The paper introduces EMBER, a spectral imaging method using machine learning to identify amyloid strains in neurodegenerative diseases.