Protein Structure and Dynamics

GalaxyMemMD provides a reproducible, high-throughput pipeline for membrane protein simulation and analysis, integrating embedding, MD simulation, and downstream analysis.

AbRefine improves antibody CDR design by fusing structural and sequence predictions using framework context, achieving better recovery and lower log-likelihood than static fusion.

This study reveals a dimer-seeded nucleation mechanism for ProTα-H1 biomolecular condensates using simulations and shows how electrostatic screening and protein supersaturation control nucleation and condensate properties.

EnsPlex improves protein complex structure prediction by integrating multi-source sampling and topology-aware candidate selection, achieving a 27.8% relative improvement over AlphaFold3 in antigen-antibody systems.

The paper analyzes the evolution of correlated conformational dynamics in influenza A virus NS1 proteins, revealing a core-periphery structure that enables evolutionary sequence variation while maintaining structural robustness.

ACER improves protein-ligand co-folding by exploring alternative binding pockets and enhancing pose accuracy without retraining.

This paper introduces DeepUMQA-Selection, a framework for efficiently selecting high-quality protein complex models from large candidate pools, showing improved performance and speed.

The paper predicts quaternary structures of viral proteins using AlphaFold2 and AlphaFold-Multimer, identifying new insights into vaccinology and protease targets.

The paper introduces SOP-Multi-π, a coarse-grained model with rescaled aromatic interactions to better simulate IDPs and their phase behavior in condensates.

SMartini automates the creation of Martini 3 force field parameters for small molecules using molecular structures or SMILES strings, validated on drug-like compounds and metabolites.

This paper introduces a multi-chain Hamiltonian replica exchange framework to improve sampling of protein conformations, demonstrating its effectiveness in analyzing transient helices and order-disorder transitions in the IM30 protein.

This paper introduces HBDesigner, a deep learning tool for designing hydrogen-bond networks in proteins, showing improved performance over existing methods.

The paper investigates the functional redundancy of stickers and spacers in the intrinsically disordered region of Canoe, revealing their roles in AJ function and robustness.

This study examines the oligomerisation of AFP6, an antifreeze peptide from winter flounder, using structural prediction and biochemical assays, finding evidence for monomers, dimers, and larger oligomers.

cryoWEight integrates cryo-EM data with weighted ensemble simulations to generate conformational ensembles of biomolecules.

The paper develops a scaffold-constrained generative model for protein design, showing it produces high-quality candidates with better ranking scores than previous methods.

GuideFlip co-designs protein structure and sequence for flexible targets, improving in silico success rates and experimentally validating binders to disordered proteins.

This study evaluates how well three structural alphabets retain fold-level information compared to full-structure comparisons using TM-scores as a reference.

This study modifies de novo α-helical barrel biocatalysts using deep learning and computational methods to improve solubility and catalytic performance while maintaining structural stability.

This study reveals the molecular mechanism of nirmatrelvir binding to SARS-CoV-2 main protease using combined NMR titrations and weighted ensemble simulations, identifying two distinct binding pathways and key residues involved in resistance.

OmegaTherm predicts mutation-induced protein stability responses by jointly modeling folding free energy and melting temperature, achieving state-of-the-art performance in prediction and variant ranking.

MuseDrift generates protein variants with specified sequence identity to a wild-type sequence, achieving higher structural prediction confidence and better mutation-effect scores than existing methods.

This paper converts the N-terminal bundle of Apolipoprotein E to be membrane-compatible using reverse-QTY substitutions without altering its fold, demonstrating preserved helix content and increased hydrophobicity.

This study extends the osmophobic effect to protein side chains by developing a transfer model that accounts for backbone-side chain interactions, revealing different contributions of backbone and side chains to protein stabilization by various osmolytes.

The paper develops a multiscale model to study how conformational dynamics regulate the PP1-GADD34-eIF2 complex, showing folding-dependent assembly enhances biochemical function.

AtomWeaver enables non-canonical peptide design by using a geometric prior to generate side-chain atoms without restricting residue identity, achieving high agreement with experimental results.

ProDive detects cross-family local conservation using a pHMM-based framework, identifying 318,289 high confidence segment correspondences across 25,545 Pfam families.

AF3 Inspector provides an interactive web tool to visualize and analyze all AlphaFold 3 structural models, addressing limitations in the official service.

The paper shows targeted finetuning can enable co-folding models to learn ligand-induced conformational states not present in training data.

This paper assesses the accuracy of different force field combinations for simulating intrinsically disordered proteins and finds that combining C36m with OPC water improves accuracy for IDPs without compromising globular protein simulations.