Protein Structure and Dynamics

The paper uses simulations to analyze residue-level contact networks in IDRs, showing how phosphorylation rewires these networks and how force fields influence residue behavior.

The paper benchmarks molecular methods for protein engineering and evaluates AI-based designs by comparing mutagenesis and phenotypic screening of a fluorophore.

PRIME-ISP predicts interaction sites across diverse biomolecular interactions using a hierarchical structural framework, outperforming sequence-based methods and matching specialized approaches without task-specific features.

The paper introduces DEEP-Boltz, a framework that extracts energetic potentials from protein structure predictors, outperforming existing methods in predicting mutation-induced stability changes.

This study elucidates the mechanism of gelsolin activation through a combination of experimental and computational methods, revealing a dynamic, domain-dependent process leading to a heterogeneous active-state ensemble.

This paper presents a rapid, automation-friendly pipeline for characterizing de novo protein binders without cell culture or purification, achieving binding affinity and kinetics in three days.

Chaperone, a DataFrame-based framework and agent harness for protein design tools, cut tokens 86% and GPU-hours 64% while raising pass rate from 6% to 12%.

This study identifies two distinct portals for ligand transport in CYP17A1 using molecular dynamics and experimental validation, showing that both are required for catalytic activity.

This paper explores sequence constraints for de novo 310-helical bundles by mutating key residues in a designed peptide and assessing stability and functionality.

N-Zinclopedia predicts transition metal-binding sites and candidate metalloenzymes across protein folds, expanding the known landscape of metal utilization.

Distance-AF Multimer improves protein complex structure prediction by incorporating user-defined residue distance constraints, achieving competitive or better performance than existing methods.

This study uses molecular dynamics simulations to compare the conformational energetics of Ser195 and Thr195 in thrombin, revealing differences in rotameric states and their impact on catalytic geometry.

This study identifies an intrinsically disordered region (IDR) in viperin that enables rNTP selection via a clamp-and-latch mechanism, revealing a new aspect of antiviral enzyme function.

HighFold-Lasso predicts lasso peptide structures using topology-guided methods, achieving perfect topology success and low RMSD on a test set of 52 structures.

The paper applies the implicit function theorem to reparameterize molecular force fields for specific protein classes, improving simulation accuracy for amyloid fibrils.

This study investigates how CBM2 proteins achieve distinct substrate preferences through a balance of topological complementarity, hydrophobic dehydration, and hydrogen-bonding interactions.

The paper develops a continuum model for biomolecular condensation using the Cahn-Hilliard equation, validated with simulations of FUS protein phase separation.

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.

The paper investigates how structural templates affect apparent links between protein stability predictions and longevity in tetrapods, finding that template choice can create spurious associations.

The paper introduces a hybrid framework to isolate stability-independent activity effects in enzymes using dual-graph deconvolution, validated with experimental data.

This study uses molecular dynamics simulations and Markov state models to compare activation mechanisms of tyrosine kinases, revealing kinase-specific conformational ensembles.

The paper designs functional disordered regions in a transcription factor using biophysical principles to predict chromatin binding function.

This paper introduces edge-aware graph attention networks to interpret biophysical mechanisms from molecular dynamics simulations, showing improved prediction and communication network analysis.

Adding Boltz2 distogram edges and features to protein and RNA graph networks beats MD-correlation graphs on binding tasks and improves stability, interaction and RNA predictions without simulation.

The paper reports that two BindCraft2 campaigns produced no binders despite high interface confidence, identifying issues with confidence metrics and a disabled feature.

This study designs metamorphic proteins that switch between multiple conformations and shows how mutations can shift their conformational equilibrium.

The paper shows that intrinsically disordered regions in Drosophila Hox proteins affect the usage of conserved motifs near the homeodomain through context-dependent intramolecular interactions.

This paper presents a hybrid quantum-classical framework for peptide folding, demonstrating a workflow that outperforms quantum refinement in structural accuracy for a short peptide.

The paper extends Ramachandran diagrams by modeling amino acid and pair torsion-angle distributions conditioned on secondary structure elements, enabling more accurate conformational predictions.

This study evaluates LLMs' ability to predict mutation effects based on structural data without using dedicated predictors, finding comparable accuracy to existing methods.