Protein Structure and Dynamics

This paper introduces StructureDCA, a method that improves mutational landscape predictions by integrating structural information into direct coupling analysis.

PySteMoDA provides an automated tool for analyzing steered molecular dynamics simulations, improving accuracy and accessibility in computational biochemistry and biophysics.

GYDE provides an open-source platform for bench scientists to access AI tools for protein design and drug discovery through a collaborative interface.

This study uses holographic imaging to show that Bovine Serum Albumin phase separation depends on higher-order oligomers and is tunable by pH.

ProTact improves protein-protein interaction prediction using a geometric graph neural network with physics-informed constraints, outperforming existing methods on benchmark datasets.

The paper investigates whether consecutive enzymes in metabolic pathways are spatially optimized for catalytic site proximity, finding no evidence for such optimization.

This study investigates how different crowding agents affect protein phase separation using NMR, revealing residue-level environmental influences and highlighting the importance of crowder identity.

The paper introduces CoMPLip, a method that improves membrane-protein structure prediction by co-folding proteins with lipid molecules during AlphaFold 3 predictions, enhancing ligand-pose prediction and conformational sampling.

The paper benchmarks four AI tools for designing BMPR1A-binding peptides, identifies top candidates, and highlights differences in binding site targeting.

ProteinMCP accelerates and democratizes protein engineering by automating workflows and integrating 38 tools, achieving a comprehensive fitness modeling task in 11 minutes.

CombinGym benchmarks machine learning for combinatorial protein variant design using curated datasets and evaluates model performance across multiple tasks and data splits.

The paper identifies and analyzes human intrinsically disordered regions (IDRs) in transcription factors, revealing their unique functional, phenotypic, and evolutionary characteristics compared to non-TF IDRs.

NEAT-DNA is a coarse-grained DNA model that improves physical realism and sequence specificity for large-scale simulations of genome structure.

This study uses explainable AI to analyze the energy landscape of protein kinases, revealing that allosteric sites are in neutrally frustrated regions, explaining their detection challenges.

The paper shows that mutational robustness correlates with protein dynamics across natural and designed proteins, outperforming existing structural confidence scores.

This study benchmarks deep learning methods for predicting antibody-antigen complex structures, finding that increased sampling improves accuracy but model selection remains challenging, with AlphaFold3 performing best overall.

This study uses molecular dynamics simulations to show that the FUS RGG3 IDR forms a dynamically organized dense phase with defined interaction motifs and multi-scale topology.

This paper introduces a coarse-grained simulation method to predict the viscosity of antibody solutions and their response to excipients.

ST-PARM improves multi-objective protein design by aligning a frozen language model with user-specified trade-offs using a reward model that handles uncertainty and achieves broader Pareto coverage.

This paper introduces Martini3-NMR, a framework that integrates NMR data into coarse-grained simulations to improve the accuracy of protein conformational ensembles.

evedesign provides a unified framework for conditional biosequence design, enabling multiobjective workflows and experimental integration across various applications.

StrucTTY provides real-time, interactive visualization of protein structures in text-only terminals using ASCII graphics and supports multi-structure comparison and Foldseek alignments.

DynMoCo identifies modular substructures in protein dynamics from molecular simulations by detecting time-evolving communities of residues.

This study reveals that ezrin activation involves spontaneous FERM-CTD dissociation triggered by PI(4,5)P2 binding, with phosphorylation stabilizing the open conformation.

The paper shows that intrinsically disordered regions in proteins amplify DNA structural fluctuations, enabling allostery in transcription factor binding.

drFrankenstein automates the generation of AMBER force field parameters for non-canonical amino acids, improving the efficiency and accuracy of molecular dynamics simulations.

This paper introduces ProteinEBM, a protein conformational energy model that outperforms existing methods in predicting mutation effects and protein structures.

The paper shows that function can be conserved in transcriptional activation domains without sequence conservation through evolutionary turnover of key residues.

The paper accelerates protein structure prediction using OpenFold with TensorRT and MMseqs2-GPU, achieving 131x faster inference than AlphaFold2 on specific hardware.

AlphaSAXS uses SAXS data to guide AI in reconstructing accurate protein conformations, improving upon sequence-only models by incorporating experimental constraints.