Protein Structure and Dynamics

This study investigates how pathogenic BRCA1 mutations disrupt BARD1-mediated allosteric control, revealing conformational changes that impair E3 ligase activity and contribute to cancer.

The paper identifies a 9-nucleotide deletion in honey bee Vg alleles and shows it maintains protein structure and stability using structural bioinformatics and a transformer-based model.

The paper investigates the zinc-induced folding and solution structure of the ZC4H2 zinc finger, showing it folds around zinc with picomolar affinity and has a novel structural motif.

This study uses reverse evolution simulations to show that amino acid alphabets evolved to improve protein folding, but early alphabets may not have favored deeply knotted structures.

This study investigates how ligand binding affects the dynamics of the bacterial DNA sliding clamp beta-clamp using NMR and hydrogen-deuterium exchange mass spectrometry, revealing dynamic allosteric effects.

EnsembleFlex provides a tool for dual-scale flexibility analysis of protein structure ensembles, combining experimental and predictive methods for backbone and side-chain dynamics.

This paper introduces DMS-Fold, a method that uses deep mutational scanning data to improve AlphaFold2 predictions, achieving better results for most protein targets.

The paper investigates the aggregation propensity of the signal peptide of Human Serum Albumin (residues 1-18) and shows it forms amyloid-like fibrils.

The paper evaluates the stability and function of randomized hydrophobic protein cores across evolutionary distances, finding that allostery disrupts function despite stable cores.

This study identifies a conserved allosteric network in apoptotic caspases across chordates, revealing how conformational dynamics and residue interactions regulate their function over 500 million years of evolution.

ActSeek searches the Alphafold database for proteins with active sites similar to a seed protein, identifying enzymes for biodegradable plastics and drug off-targets.

This study investigates how phosphorylation of ABLIM1's intrinsically disordered regions modulates its interaction with titin, affecting cardiac muscle contraction.

This paper introduces NIPAD, a method to design two-component tubular protein assemblies that mimic biological structures like actin and microtubules, showing dynamic flexibility and reversible assembly.

This study calculates amino acid transfer free energies to reveal thermodynamic contributions to biomolecular condensate formation, showing charge asymmetry and interactions between protein and water.

This study evaluates ML architectures for protein fitness prediction using simulated and empirical landscapes, identifying landscape ruggedness as a key determinant of accuracy.

This paper introduces RASP, a deep learning model that improves structure prediction with restraints, and FAAST, a pipeline that accelerates NMR NOESY peak assignment and structure ensemble generation.

SOLeNNoID is a deep learning pipeline that detects solenoid residues in protein structures, achieving a 71% increase in solenoid-containing entries compared to RepeatsDB.

SeaMoon predicts continuous protein motions from sequences using pLM embeddings and a lightweight CNN, achieving reasonable accuracy for 40% of test proteins.

AF3Score is a score-only adaptation of AlphaFold3 that evaluates biomolecular structures without iterative refinement, showing improved performance in binder screening and fold-switching protein analysis.

This paper introduces a computational method to identify allosteric networks in proteins by analyzing hydrogen bond fluctuations in the apo state, demonstrating its effectiveness across three proteins.

DSSP 4 annotates protein secondary structure in a FAIR-compliant manner, introducing left-handed κ-helices and improving interoperability with mmCIF format.

The paper designs proteins that bind intrinsically disordered regions with high affinity and specificity across diverse targets.

This study investigates how phosphorylation affects the conformational dynamics of the disordered GR NTD, revealing distinct topological changes and introducing a new analysis framework.

This paper develops a coarse-grained model for PEG to study crowding effects on disordered proteins, showing how PEG titrations can quantify phase separation propensities.

The paper evaluates AI and first-principle methods for protein redesign, showing hybrid approaches are most reliable, while inverse folding tools and force fields have specific strengths and limitations.

This paper designs a water-soluble CD20 mimic using computational methods to enable binding studies and therapeutic development.

The paper introduces OpenCafeMol, a GPU-accelerated MD simulator for biomolecules, showing speedups over CPU and applying it to vesicle fusion studies.

This study uses molecular dynamics simulations to analyze the dynamic behavior of S. aureus Type I signal peptidases SpsA and SpsB, including the impact of the P29S mutation on their structure and function.

This study compares how polyphosphate and DNA influence the assembly of disordered proteins, showing polyP's superior ability to distinguish protein conformations and drive diverse assembly behaviors.

This study identifies that increased flexibility, not compactness, underlies thermal adaptation in prokaryotic adenylate kinases across 70 species.