Frank DiMaio

Complex enzyme active site scaffolding by iteratively detuned catalytic guidance

CaGE alternates guided structure prediction with sequence redesign, fading guidance each cycle, so sequences recover complex catalytic sites unaided; designed nonheme-iron and PLP-dependent enzymes were active.

RFOptimization: Guiding Design Optimization with All-Atom Structure Prediction

RFOptimization improves all-atom biomolecular binder design by combining gradient-guided sequence optimization with orthogonal predictors to enhance in silico success rates.

COPAL: An ensemble of protein-ligand co-folding models enriches preferred ligand predictions for the LuxR-family of quorum sensing receptors

COPAL uses an ensemble of protein-ligand co-folding models to predict preferred AHL ligands for LuxR receptors, achieving higher accuracy than individual models.

Linker-Length Landscape Mapping Enables Coupling of Diverse Synthetic Chemically Induced Dimerization Systems to Molecular Readouts

The paper introduces a method to map linker-length landscapes for coupling synthetic CID systems to molecular readouts, showing linker length strongly affects coupling efficiency.

MX2 Mediates Collapse of the HIV-1 Capsid

This study identifies the minimal MX2 region required for HIV-1 restriction and shows that MX2 binding disrupts the capsid lattice, leading to premature fragmentation and preventing viral nuclear import.

Using experimental results of protein design to guide biomolecular energy-function development

The paper uses experimental results from de novo protein designs to identify and resolve failure modes in the Rosetta energy function, improving its accuracy in predicting stable protein structures.

Modeling protein-small molecule conformational ensembles with PLACER

An atom-level graph neural network that denoises corrupted structures generates small-molecule and protein-ligand conformational ensembles, and its active-site pre-organization metric enriched for more active designed retroaldolases.

De novo design of RNA and nucleoprotein complexes

This paper introduces a generative diffusion model for de novo design of RNA and nucleoprotein complexes, validated experimentally with SHAPE-seq and electron microscopy.

De novo design of phosphotyrosine peptide binders

This paper introduces RFD2-MI, a deep learning framework for designing phosphotyrosine peptide binders, showing high specificity and affinity for phosphorylated targets.

De novo Design of All-atom Biomolecular Interactions with RFdiffusion3

An all-atom diffusion model using a 14-atom residue representation and sparse U-Net architecture accelerates biomolecular design 10-fold, enabling experimentally validated DNA-binding proteins and multi-turnover cysteine hydrolases.

Computational design of sequence-specific DNA-binding proteins

This paper designs sequence-specific DNA-binding proteins computationally and validates them experimentally, achieving high specificity and affinity for target sequences.

Accelerating Biomolecular Modeling with AtomWorks andRF3

AtomWorks, a modular atom-level data framework, trains structure prediction and design models; the resulting RF3 improves chirality handling and performs between AF3 and Boltz.

Automated identification of small molecules in cryo-electron microscopy data with density- and energy-guided evaluation

This paper introduces EMERALD-ID, a method for identifying small molecules in cryo-EM data, achieving 43% exact identification and 66% close identification of common ligands.

Accurate de novo design of high-affinity protein binding macrocycles using deep learning

This paper introduces RFpeptides, a deep learning framework for designing high-affinity macrocyclic peptides, achieving sub-nanomolar binding affinities and accurate structural predictions.

Recognition of unique cell surface glycopolymers by diversified lectin domains specifies umbrella toxin targeting

The paper identifies teichuronic acid oligosaccharides as the receptor for umbrella toxins and shows targeting specificity arises from lectin domain recognition of variable TUA-WTA structures.

An artificial intelligence accelerated virtual screening platform for drug discovery

The paper develops RosettaVS, an AI-accelerated virtual screening platform that identifies drug candidates with high binding affinity across two targets, achieving a 14% and 44% hit rate in under seven days.

Towards high-resolution modeling of small molecule-ion channel interactions

This study evaluates two ligand docking methods for ion channel complexes, finding they can sample near-native poses but struggle to consistently identify them as top-scoring models.

Broadly applicable and accurate protein design by integrating structure prediction networks and diffusion generative models

Fine-tuning RoseTTAFold as a diffusion denoiser gives RFdiffusion, a general protein backbone generator; a designed influenza haemagglutinin binder matched its model by cryo-EM.

Accurate prediction of nucleic acid and protein-nucleic acid complexes using RoseTTAFoldNA

One RoseTTAFold-derived network models protein-DNA and protein-RNA complexes plus RNA tertiary structure end to end, returning confidence estimates alongside each 3D model.

Seipin forms a flexible cage at lipid droplet formation sites

This paper reports the structure of yeast seipin, revealing a decameric cage that facilitates lipid droplet formation through conformational changes.

Genome sequencing of 196 Treponema pallidum strains from six continents reveals additional variability in vaccine candidate genes and dominance of Nichols clade strains in Madagascar

Ensuring scientific reproducibility in bio-macromolecular modeling via extensive, automated benchmarks

The paper presents a framework for ensuring reproducibility in bio-macromolecular modeling through automated benchmarks and a test server, demonstrating its utility for Rosetta and other tools.

Growing Glycans in Rosetta: Accurate de novo glycan modeling, density fitting, and rational sequon design

This paper expands the Rosetta software suite with tools for glycan modeling, density fitting, and rational sequon design, achieving high accuracy in predicting and building glycan structures.

Accurate prediction of protein structures and interactions using a 3-track network

The paper introduces a 3-track network for accurate protein structure and interaction prediction, achieving performance close to DeepMind's CASP14 results and enabling rapid modeling of protein complexes from sequence data.

Structural control for the coordinated assembly into functional pathogenic type-3 secretion systems

This study reveals the structural control of type-3 secretion system assembly in Salmonella, showing how structural features guide the formation of a 24:16/15 basal body and regulate protein secretion.

3.1A structure of yeast RNA polymerase II elongation complex stalled at a cyclobutane pyrimidine dimer lesion solved using streptavidin affinity grids

The paper solves a high-resolution structure of a stalled RNA polymerase II elongation complex using streptavidin affinity grids to overcome air-water interface interactions.

Atomic model of microtubule-bound tau

The paper uses cryo-EM to determine the atomic structure of tau-tubulin interactions, revealing how tau stabilizes microtubules through specific binding sites.

An integrated, pipeline-based approach for cryo-EM structure determination and atomic model refinement in the cloud

The paper develops cloud-based software tools for cryo-EM structure determination and atomic model refinement, reducing barriers for new users.