diffusion model

Vesicular acidification modulates the synaptic current: a hybrid diffusion reaction model analysis

The paper develops a diffusion-reaction model to analyze how vesicular acidification modulates AMPA receptor function and synaptic current.

Cross-modal Denoising and Integration of Spatial Multi-omics data with CANDIES

CANDIES denoises and integrates spatial multi-omics data using a conditional diffusion model and contrastive learning, improving downstream analysis tasks.

HMCVelo: A Deterministic Model for Hydroxymethylation Velocity in Single Cells

HMCVelo computes hydroxymethylation velocity in single cells using a deterministic ODE model, outperforming RNA velocity in trajectory inference for murine cortical cells.

Modular Deep Learning for Direct RNA Sequence Design via Self-Contained RNA Units

The paper introduces SCRU-DB, a database of RNA units, and two models, SCRU-Seq and SCRU-Diff, to design RNA sequences with high structural accuracy.

Virtual multiplex staining of the pancreatic islets across type 1 diabetes progression using a Schroedinger bridge

This paper introduces SMILE, a diffusion model for converting H&E stains to multiplex IHC images, showing superior performance in pancreatic islet analysis and breast cancer stain conversion.

GANGE: Achieving Sequencing Without Sequencing With Diffusion Guided Generative Genomic Transformer

GANGE is a deep-learning system that generates accurate genomic sequences from low-coverage, error-prone reads, reducing sequencing costs and enabling regulomics research without genome data.

A generative AI framework for disease-specific lung microtissue bioengineering

This paper introduces a generative AI framework for designing and fabricating disease-specific lung microtissues using 3D imaging and bioprinting.

Coherent Cross-modal Generation of Synthetic Biomedical Data to Advance Multimodal Precision Medicine

The paper develops a generative framework to synthesize missing modalities in multimodal datasets, validated on cancer data, showing it can maintain predictive model performance on incomplete profiles.

How Generative Models Approach Molecular Conformational Sampling

This study compares diffusion and rectified-flow models in generating molecular conformations, showing diffusion models better capture ensemble structure while rectified-flow requires more expressive architectures.

High-PepBinder: A pLM-Guided Latent Diffusion Framework for Affinity-Aware Target-Specific Peptide Design

High-PepBinder generates target-specific peptides using a pLM-guided diffusion model and a comprehensive dataset, producing peptides with favorable binding properties and sequence diversity.

Generative cerebral vasculature visualization using spatialtranscriptomic data

The paper uses spatial transcriptomic data and a generative model to predict brain vasculature organization from gene co-expression patterns.

Antidepressants interact with an LPA3 receptor binding site: functional and docking studies

This study investigates how antidepressants interact with LPA3 receptors through functional assays and docking simulations, revealing distinct effects on calcium and ERK 1/2 signaling.

Scaling and Generalization of Discrete Diffusion Models for Tumor Phylogenies

This study uses discrete diffusion models to generate tumor phylogenies, finding that mid-scale models outperform deeper ones and that diverse training improves cross-regime generalization.

Computational Design and Atomistic Validation of a High-Affinity VHH Nanobody Targeting the PI/RuvC Interface of Streptococcus pyogenes Cas9: A Bivalent Hub Strategy for CRISPR-Cas9 Enhancement

The paper computationally designs a VHH nanobody targeting Cas9's PI/RuvC interface and validates its structure and stability through molecular dynamics simulations.

X-Cell: Scaling Causal Perturbation Prediction Across Diverse Cellular Contexts via Diffusion Language Models

X-Cell predicts causal perturbation responses across diverse cellular contexts using a diffusion language model trained on a large CRISPRi Perturb-seq dataset.

ChironRNA: Steric Clashes Resolution in RNA Structures via E(3)-Equivariant Diffusion

ChironRNA resolves steric clashes and reconstructs missing atoms in RNA structures using an E(3)-equivariant diffusion model, achieving high precision on test data.

Single-Pass Discrete Diffusion Predicts High-Affinity Peptide Binders at >1,000 Sequences per Second across 150 Receptor Targets

LigandForge generates high-affinity peptide binders rapidly across multiple receptor targets using a discrete diffusion model, outperforming existing methods in speed and binding quality.

Constrained Diffusion as a Paradigm for Evolution

This paper introduces DiffEvol, a constrained diffusion framework to model evolution, and applies it to SARS-CoV-2 data to reconstruct fitness trends and vaccine-induced phase transitions.

Multiscale conformational sampling of multidomain fusion proteins by a physics informed diffusion model

This paper introduces a physics-informed diffusion model to efficiently sample conformational ensembles of multidomain fusion proteins, outperforming traditional methods in capturing large-scale dynamics.

Automated high-throughput fabrication of patient-specific vessel-on-chips enables a generative AI digital twin--Cascade Learner of Thrombosis (CLoT) for personalized thrombosis prediction

The paper develops a platform for automated vessel-on-chip fabrication and a generative AI model, CLoT, to predict thrombosis in patients with high accuracy.

CellPace: A temporal diffusion-forcing framework for simulation, interpolation and forecasting of single-cell dynamics

CellPace simulates, interpolates, and forecasts single-cell developmental dynamics using a temporal diffusion model with continuous time encoding, achieving state-of-the-art performance on mouse data.

Conditional Diffusion with Locality-Aware Modal Alignment for Generating Diverse Protein Conformational Ensembles

This paper introduces Mac-Diff, a diffusion model that generates diverse protein conformations by aligning sequence and geometry views, showing improved recovery of conformational distributions and meta-stable states.

Physiologically realistic gamma activity produced in silico by weakening the PING attractor state

This paper modifies the PING mechanism to produce more realistic gamma oscillations by introducing biophysically motivated changes that reduce synchronization and increase transient activity.

Discrete Diffusion for Single-Cell Gene Expression Modeling

This paper introduces Discrete Cell Models (DCM), a diffusion-based framework for single-cell gene expression modeling that operates in the discrete domain, achieving strong performance on benchmark datasets.

Diffusion Probabilistic Models for Missing-Wedge Correction in Cryo-Electron Tomography

The paper introduces MW-RaMViD, a diffusion model for correcting missing-wedge artifacts in cryo-ET by generating missing tilt images, showing improved reconstruction fidelity with smaller step sizes and larger conditioning windows.

Metadiffusion: inference-time meta-energy biasing of biomolecular diffusion models

Metadiffusion introduces a meta-energy biasing layer to diffusion models to generate diverse biomolecular conformations without retraining, matching molecular dynamics simulations.

BESTish: A Diffusion-Approximation Framework for Inferring Selection and Mutation in Clonal Hematopoiesis

BESTish infers mutation fitness, mutation rate, and environmental strength in clonal hematopoiesis using a diffusion-approximation framework applied to VAF datasets.

Ophiuchus-Ab: A Versatile Generative Foundation Model for Advanced Antibody-Based Immunotherapy

Ophiuchus-Ab is a generative foundation model for antibody design that captures inter-chain dependencies and predicts antibody properties like developability and binding affinity.

Designing protein/non-protein binding interactions using a full-atom diffusion model

The paper introduces AtomPaint, a full-atom diffusion model for designing protein/non-protein binding interactions, showing it outperforms random guessing in sequence recovery.

Modelling Immune Dynamics in Locally Advanced MSI-H/dMMR Colorectal Cancer with Neoadjuvant Pembrolizumab Treatment: From Differential Equations to an Agent-Based Framework

This paper develops a differential equation model and converts it to an agent-based model to simulate immune dynamics in MSI-H/dMMR colorectal cancer under pembrolizumab treatment.