diffusion model

DiffusionST: A deep generative diffusion model-based framework for enhancing spatial transcriptomics data quality and identifying spatial domains

DiffusionST improves spatial transcriptomics data quality and identifies spatial domains through a diffusion model and graph convolutional network.

ExEnDiff: An Experiment-guided Diffusion model for protein conformational Ensemble generation

ExEnDiff generates protein conformations using experimental data as a prior, improving diversity and physical realism over existing methods.

Cycledesigner Leveraging RFdiffusion and HighFold to Design Cyclic Peptide Binders for Specific Targets

This study introduces cycledesigner, a method combining RFdiffusion, ProteinMPNN, and HighFold to design cyclic peptide binders for specific targets, followed by scoring functions for screening.

Homeorhetic regulation of cellular phenotype

The paper introduces a framework called signaling homeorhesis to explain how cells interpret growth factor signals over time to produce distinct phenotypes.

A conditional generative model to disentangle morphological variation from batch effects in model organism imaging studies

This paper introduces a conditional latent diffusion model to separate batch effects from morphological features in zebrafish imaging data, enabling accurate genotype classification.

Distinct interferon response patterns link to increased transposable element expression in leukocytes of systemic lupus erythematosus patients

This study links increased transposable element (TE) expression in SLE patients to IFN responses, showing strong correlations with disease activity and splicing alterations.

A Diffusion-Based Framework for Designing Molecules in Flexible Protein Pockets

YuelDesign generates molecules for flexible protein pockets using a diffusion framework, producing drug-like molecules with diverse functional groups and comparable docking energies to native ligands.

DynaRNA: Dynamic RNA Conformation Ensemble Generation with Diffusion Model

DynaRNA generates RNA conformation ensembles using a diffusion model with EGNN, outperforming MD simulations in accuracy and efficiency.

LLPS REDIFINE allows the biophysical characterization of multicomponent condensates without tags or labels

The paper introduces REDIFINE, a label-free method to characterize multicomponent biomolecular condensates using diffusion NMR, revealing insights into their biophysical properties and dynamics.

Poxvirus targeted by RFdiffusion peptide-binders

The paper uses RFdiffusion to generate peptide-binders targeting poxvirus F13L homodimer interfaces, achieving picoMolar affinities and improved stability against proteases.

SynLS: A novel diffusion-transformer framework for generating high-quality wearable sensor time series data to enhance health monitoring

SynLS generates synthetic wearable sensor data to improve health monitoring, enhancing prediction tasks by 18.5% and 26.8% in livestock health events.

MedSAGE: Bridging Generative AI and Medicinal Chemistry for Structure-Based Design of Small Molecule Drugs

MedSAGE improves small molecule drug design by generating synthesizable, drug-like molecules with high predicted affinity and selectivity for therapeutic targets.

GeoFlow-V2: A Unified Atomic Diffusion Model for Protein Structure Prediction and De Novo Design

GeoFlow-V2 is a unified atomic diffusion model for protein structure prediction and de novo design, showing strong performance in both tasks and offering a web server for accessibility.

FKSFold: Improving AlphaFold3-Type Predictions of Molecular Glue-Induced Ternary Complexes with Feynman-Kac-Steered Diffusion

Feynman-Kac steering with an ipTM reward at inference time lets Chai-1r recover three of eight molecular-glue ternary complexes below 3 Å without retraining.

Ivermectin binds to the allosteric site (site 2) and inhibits allosteric integrin activation by pro-inflammatory cytokines.

This study shows ivermectin binds to the allosteric site 2 of integrins and inhibits their activation by pro-inflammatory cytokines, similar to anti-inflammatory cytokines.

Robust and Reliable de novo Protein Design: A Flow-Matching-Based Protein Generative Model Achieves Remarkably High Success Rates

The paper introduces OriginFlow, a protein generative model that achieves high success rates in designability, diversity, and functional design of proteins, validated through wet-lab experiments.

Fast and Accurate Peptide - MHC Structure Prediction via an Equivariant Diffusion Model

MHC-Diff predicts peptide-MHC structures with high accuracy and speed using a diffusion model, outperforming existing methods.

PhysDock: A Physics-Guided All-Atom Diffusion Model for Protein-Ligand Complex Prediction

PhysDock predicts protein-ligand complexes using a physics-guided diffusion model, achieving state-of-the-art performance in redocking and cross-docking benchmarks.

DiffPIE: Guiding Deep Generative Models to Explore Protein Conformations under External Interactions

This paper introduces DiffPIE, a physics-guided method to predict protein conformations under external interactions without retraining, showing success in cyclic peptides and gold-surface adsorption.

Degeneracy in negative feedback (NFBL) and incoherent feedforward (IFFL) loops: Adaptation and resonance

The paper analyzes degeneracy in adaptive circuit mechanisms, showing how it leads to adaptation and resonance in response to different input types.

ARTEMIS integrates autoencoders and schrodinger bridges to predict continuous dynamics of gene expression, cell population and perturbation from time-series single-cell data

FMed-Diffusion Federated Learning on Medical Image Diffusion

This paper proposes FMedDiffusion, a federated learning approach for medical image diffusion models, achieving performance comparable to traditional training on three medical image datasets.

MolSnapper: Conditioning Diffusion for Structure Based Drug Design

MolSnapper improves structure-based drug design by conditioning diffusion models with 3D pharmacophores, generating more valid molecules with higher structural and chemical similarity.

Versatile Stain Transfer in Histopathology Using a Unified Diffusion Framework

This paper introduces DUST, a diffusion model for histopathology stain transfer, showing improved performance over GAN-based methods on a kidney histopathological dataset.

DTMol: Pocket-based Molecular Docking using Diffusion Transformers

DTMol is a diffusion model for pocket-based molecular docking that outperforms existing methods on benchmark datasets and identifies new JAK2 inhibitors.

Evaluating Synthetic Diffusion MRI Maps created with Diffusion Denoising Probabilistic Models

The paper generates synthetic diffusion MRI maps using diffusion models and evaluates their realism and utility for improving AI classifiers.

De novo peptide sequencing with InstaNovo: Accurate, database-free peptide identification for large scale proteomics experiments

InstaNovo is a transformer model for de novo peptide sequencing that outperforms existing methods, enabling accurate identification of peptides and discovery of novel sequences and organisms.

Ocelli: an open-source tool for the visualization of developmental multimodal single-cell data

Ocelli is a Python tool for visualizing developmental multimodal single-cell data using diffusion-based modeling, improving computational efficiency and low-dimensional representation quality.

Excitability and travelling waves in renewable active matter

This study investigates mechanical excitability in renewable active matter, showing that activity and turnover generate travelling waves and spatiotemporal chaos in systems like the cell cytoskeleton.

Combining transformer and 3DCNN models to achieve co-design of structures and sequences of antibodies in a diffusional manner

The paper introduces AlphaPanda, an algorithm combining transformer, 3DCNN, and diffusion models for antibody sequence and structure design, achieving high-quality results with reduced training data.