Yang Zhang

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.

RNArefine improves RNA structure prediction by refining atomic details and stereochemistry using AI and physics-based methods.

This paper introduces a protein-protein language model for predicting interactions, binding affinity, and interface contacts, achieving state-of-the-art results across species and complex systems.

ProteinConformers provides a large-scale dataset of protein conformations with energy profiles and similarity annotations, offering a benchmarking framework and analysis platform.

AlphaFlex improves prediction of multiple protein conformations by integrating flexible residue prediction and directed masking, outperforming existing methods on apo-holo pairs.

MEBOCOST infers metabolite-mediated cell communications from single-cell RNA-seq data, identifying known and novel interactions in mouse brown adipose tissue.

This paper introduces Ouroboros, a framework that improves drug discovery by integrating molecular representation learning with generative modeling, outperforming traditional methods in multiple tasks.

The paper uses multivariate coevolutionary analysis to show how life-history traits evolve through modular pathways in amniotes, with distinct tempos across traits and taxa.

DRfold2 improves RNA structure prediction using a composite language model and denoising module, outperforming existing methods in contact prediction and complementing AlphaFold3.

This study evaluates six deep learning models for RNA 3D structure prediction across multiple datasets, identifying their strengths and limitations.

ModelCIF extends the PDBx/mmCIF standard to represent predicted 3D protein models, enabling FAIR data sharing for computational structural biology.

This paper introduces FoldDesign, a method for de novo protein fold design that creates novel structures from secondary structure assignments, achieving high fidelity and stability.

This paper introduces DRfold, a method that improves RNA tertiary structure prediction by integrating deep learning with geometric potentials, achieving significant performance gains over existing methods.

The paper investigates the evolutionary origins of myoblast fusion mechanisms in chordates, identifying the emergence of Myomaker and Myomixer and their roles in muscle development across species.

US-align provides a universal platform for aligning diverse macromolecular structures, improving accuracy and speed in structural comparisons.

The paper introduces ATGO, a deep-learning method that improves Gene Ontology prediction for proteins using self-attention transformers and triplet networks, achieving higher accuracy than existing methods.

PEPPI predicts protein-protein interactions using structural, sequence, and functional data, achieving higher accuracy than existing methods and identifying interactions relevant to SARS-CoV-2 infection.

DeepFoldRNA couples deep self-attention neural networks with gradient-based folding simulations to predict atomic-resolution RNA tertiary structures from sequence in ~1 minute with 2.69 A average RMSD.

TripletGO improves gene function annotations by integrating transcript expression, sequence alignment, and probability models across multiple species and proteins.

This paper introduces DEMO-EM, a method for assembling multi-domain protein structures from cryo-EM maps, achieving high accuracy and outperforming existing methods.

This study identifies RTL8 as a novel UBQLN2 interactor that promotes its nuclear translocation and subnuclear localization in protein quality control.

The paper introduces MPDB, a multi-domain protein structure database, and a server with modules for filtering and detecting structural analogues to aid in multi-domain protein modeling.

The paper identifies potential SARS-CoV-2 receptors, including CAT, AGTR2, L-SIGN, and DC-SIGN, which show higher binding affinity to the S protein than ACE2 and are expressed in affected tissues.

This study investigates the neuronal mechanism of a BK channelopathy in absence epilepsy and movement disorders using a knock-in mouse model and finds that BK channel inhibition can mitigate seizures and motor deficits.

MMpred improves de novo protein structure prediction by using a multimodal optimization protocol that enhances conformation sampling and model accuracy.

The paper improves protein structure prediction by linking microbiome sequences to protein families using a targeted approach with predicted biomes.

TripletRes predicts high-accuracy protein contact-maps from primary sequences using deep residual convolutional networks and coevolutionary matrices, outperforming existing methods on benchmark datasets.

The paper develops a pipeline integrating multi-chain threading with high-throughput experiments to improve PPI detection and structural modeling in E. coli, achieving higher confidence PPIs and accurate complex structures.

The paper models myoglobin structures in 302 vertebrates, finds positive selection for aggregation resistance in penguin myoglobins and purifying selection in galliforms, and proposes a new equation for net charge and concentration relationship.

This study designs SARS-CoV-2 spike protein variants with enhanced immunogenicity by engineering T cell epitopes while preserving structure and function.