Protein Structure and Dynamics

Unobserved Sequence Space Has Many Functional Proteins

The paper shows that unobserved protein sequence space contains many functional proteins, challenging the assumption that existing proteins cover most functional space.

Efficient, Few-shot Directed Evolution with Energy Rank Alignment

This paper introduces a method to improve protein engineering efficiency using pre-trained protein language models and statistical physics-based post-training to generate high fitness sequences with fewer experimental data points.

Prot-LAMBDA: Explicit Distance Learning Enhances Structural Reasoning in Protein Language Models

Prot-LAMBDA improves protein structure prediction by explicitly incorporating spatial relationships, achieving better performance than ESM2-3B in contact prediction and 3D structure prediction.

Protein Surface Site Determines the Evolutionary Accessibility of Allosteric Regulation

This study shows that inserting a domain into a protein's sector-connected surface increases the likelihood of evolving allosteric regulation, while non-sector-connected sites are less evolvable.

AFP-R: An Open Resource Dedicated to Antifreeze Proteins

AFP-R provides a database and predictor for antifreeze proteins, offering curated data and a machine-learning model for AFP identification.

Emergence of new function through evolutionary divergence of an intrinsically disordered region

This study shows how transient secondary structure in intrinsically disordered regions can generate new molecular functions while preserving ancestral activities in paralogous proteins.

A discrete protein subset drives structure prediction discordance in orphan proteins

The paper identifies a subset of proteins that causes discordance between structure and disorder predictors, showing that this subset is compositionally distinct and not a general sequence-class artefact.

Spanning-Tree Thermostatistics of Protein Allostery: An Exact Kirchhoff Framework with Application to Oncogenic KRAS

Kirchhoff matrix-tree evaluation of residue-contact spanning-tree ensembles yields exact allosteric thermodynamics; KRAS G12D conserves global and channel heat capacity but reroutes paths via Q61 and F156.

Improved inference of multiscale sequence statistics in generative protein models

This paper introduces a new regularization strategy for generative protein models that improves the inference of multiscale sequence statistics, enabling the generation of functional and diverse protein sequences.

Benchmarking antibody-antigen co-folding on human monomeric antigens

This study benchmarks co-folding methods for antibody-antigen complexes using a new benchmark dataset, showing recent improvements but ongoing challenges in modeling CDRH3 loops and sampling correct binding modes.

Molecular dynamics descriptors for 1,079 post-translationally modified protein systems

The paper provides a resource of 1,079 post-translationally modified protein systems with molecular dynamics simulations and descriptors for research and method evaluation.

Thal-Kak: unifying biomolecular structure predictors reveals a sampling-selection gap

Thal-Kak combines five structure predictors to improve sampling but finds confidence scores and quality assessment methods hinder model selection.

ResiRuler: A Toolkit for Visualizing Residue-Residue Distances and Structural Changes in Biomolecular Models

ResiRuler quantifies and visualizes residue-residue distances and structural changes across protein ensembles to identify conformational differences.

Progressive Loosening of a Dual Autoinhibitory Interface Activates PP2A-B56δ

This paper investigates the activation mechanism of PP2A-B56δ by analyzing disease variant E198K using cryo-EM and computational methods, revealing progressive loosening of the N/C-arm interface as the activation pathway.

Membrane Anisotropy Reshapes Scale-Free Correlations and Directional Mechanical Susceptibility in Transmembrane Proteins

The paper analyzes transmembrane proteins under membrane constraints, finding direction-dependent correlations and new predictions for mutation-sensitive sites in GPCRs.

Autotransporter folding avoids a kinetic trap during vectorial translocation across the bacterial outer membrane

The paper explains how autotransporter proteins avoid a kinetic trap during secretion by using a computational framework to reveal hidden conformational states.

AlphaConformers: Structure-guided sampling enables prediction of multiple protein conformations

AlphaConformers improves AlphaFold2 by generating multiple protein conformations using structural information from databases, outperforming existing methods in capturing alternative states.

BindCORE: Biophysical Ensemble Learning for Predicting Interaction Sites in Intrinsically Disordered Regions

BindCORE improves prediction of interaction sites in intrinsically disordered regions by integrating biophysical ensemble features, outperforming sequence-based methods.

Morpheus-3D: Structural Diversity-Guided Detection and Localization of Protein Fold Switching

Morpheus-3D detects and localizes fold-switching proteins using sequence-based entropy profiles, outperforming existing methods and revealing widespread structural plasticity across proteomes.

Flexibility Drives Information Flow in Proteins: Fluctuation Potential Gradients Dictate Directional Entropy Transfer

The paper introduces a matrix framework to map directional entropy transfer in proteins by incorporating continuous distance-dependent weighting into the Kirchhoff matrix, applied to KRAS to identify allosteric communication pathways.

From Code to Cure: Computationally Designed BMP-2 Binders Using AI-Integrated Pipelines for Controlled Bone Regeneration

This paper designs BMP-2 binders using AI pipelines to control bone regeneration, finding a binder with 2.07 nM affinity for BMP-2.

Dynamic view of an allosteric intermediate in a positively cooperative dimer

This paper stabilizes and analyzes the singly-bound intermediate of chorismate mutase to reveal that positive cooperativity arises from dynamic ensemble behavior rather than a simple two-state switch.

Expanding the Landscape of Disordered Flexible Linkers: A Structural and Computational Framework for DLD dataset assembly

The paper introduces a classification framework for disordered flexible linkers and creates a comprehensive dataset of 1640 domain linkers, annotated with structural and sequence features.

Peptide structural plasticity is predictable from sequence and environment

ApexFold predicts peptide secondary structure changes across environments using sequence and physicochemical descriptors, outperforming static predictors.

Interaction-Range Control of Synapsin Aggregation in a Coarse-Grained Model

This paper develops a coarse-grained model of synapsin aggregation and shows that weak interaction features significantly affect collective organization at mesoscopic scales.

AFilter: Improved Antibody Epitope Prediction by Machine Learning-Optimized Interface Energy Filtering of AlphaFold3-Predicted Complex

AFilter improves antibody epitope prediction by filtering AlphaFold3 predictions using machine learning-optimized interface energy, increasing accuracy and true positive rates.

Entanglement-based continuum conformational landscape of proteins

The paper introduces mathematical topology metrics to quantify protein structure entanglement and represents the proteome in a continuous 3D space, showing structural and functional correlations.

MEM-CALVADOS: A Residue-Level Model for Flexible Proteins at Membrane Interfaces

The paper introduces MEM-CALVADOS, a residue-level model for flexible proteins at membrane interfaces, validated against experimental data and applied to signaling protein assemblies.

Protein size and geometry govern mutational robustness

This study shows that protein size, geometry, and oligomeric state systematically influence mutational robustness and evolvability.

Large-scale structure prediction of DUF-containing protein-protein interactions

A logistic regression on STRING evidence picked DUF (domain of unknown function) protein pairs for AlphaFold 3; of 12,298 modelled pairs, 1,208 met strict and 2,433 liberal confidence.