diffusion model

Angular Deviation Diffuser: A Transformer-Based Diffusion Model for Efficient Protein Conformational Ensemble Generation

This paper introduces a transformer-based diffusion model that generates protein conformational ensembles efficiently, showing alignment with MD simulations and broader conformational coverage.

EVOFLOW-RNA: GENERATING AND REPRESENTING NON-CODING RNA WITH A LANGUAGE MODEL

EvoFlow-RNA generates and represents non-coding RNA using a masked discrete diffusion model, outperforming existing models on RNA function tasks.

DrugDiff - small molecule diffusion model with flexible guidance towards molecular properties

DrugDiff generates novel small molecules with desired properties using a diffusion model guided by predictors.

Improved structural modelling of antibodies and their complexes with clustered diffusion ensembles

The paper improves antibody structure prediction by combining AlphaFlow and HADDOCK to generate diverse CDR-H3 loop conformations, enhancing docking accuracy for antibody-antigen complexes.

Synthetic Diffusion Tensor Imaging Maps Generated by 2D and 3D Probabilistic Diffusion Models: Evaluation and Applications

This study evaluates 2D and 3D diffusion models for generating synthetic DTI maps, finding 3D models perform better in downstream tasks like sex and dementia classification.

Computational design of serine hydrolases

This study designs novel serine hydrolases with high catalytic efficiency using computational methods, achieving structures close to design models and distinct from natural enzymes.

RFdiffusion Exhibits Low Success Rate in De Novo Design of Functional Protein Binders for Biochemical Detection

This study evaluates RFdiffusion's ability to design functional protein binders for six targets, finding limited success with only two Strep-Tag II binders outperforming streptavidin.

LRTM: Left-Right Transition Matrices for Molecular Interaction Prediction

This paper introduces LRTM, a method for molecular interaction prediction using transition matrices, showing improved performance and cross-task potential.

NoisyFlow: Differentially Private Optimal Transport Using Neural Networks for Secure Biomedical Data Sharing

NoisyFlow enables secure biomedical data sharing by aligning data distributions across institutions using differentially private optimal transport with neural networks.

dnaSORA - A Unified Diffusion Transformer for DNA point clouds

The paper introduces dnaSORA, a diffusion transformer for DNA point clouds, achieving state-of-the-art classification of C. elegans data at 0.3 Mb resolution using synthetic data.

A Parametric Robustness versus Dynamic Sensitivity Paradox in a Bistable Biomolecular Circuit

The paper analyzes the robustness and sensitivity of a bistable biomolecular circuit under multi-parametric perturbations, revealing a tradeoff between robustness and dynamic sensitivity.

Protenix - Advancing Structure Prediction Through a Comprehensive AlphaFold3 Reproduction

An open AlphaFold3 reproduction, with weights, inference and training code, that matches AF3 on ligand and RNA benchmarks, beats AlphaFold-Multimer 2.3 on protein interfaces and flags possible memorisation.

Conditional Denoising Diffusion Probabilistic Models with Attention for Subject-Specific Brain Network Synthesis

This paper introduces a conditional DDPM with attention to synthesize subject-specific 3D brain networks from rs-fMRI data, outperforming existing methods in accuracy and detail.

AI-Based Antibody Design Targeting Recent H5N1 Avian Influenza Strains

This study uses AI to design novel antibodies against recent H5N1 influenza strains, showing they bind similarly to known therapeutic antibodies.

Riemannian Diffusion Kernel-smoothed Continuous Structural Connectivity On Cortical Surface

This paper introduces a Riemannian diffusion kernel method for continuous structural connectivity on cortical surfaces, showing improved density estimation compared to spherical heat kernels.

RNAGenesis: A Generalist Foundation Model for Functional RNA Therapeutics

RNAGenesis is a foundation model for RNA therapeutics that integrates sequence, structure, and function prediction, achieving state-of-the-art performance on multiple tasks and enabling functional RNA design.

YDS-GlueFold: Surpassing AlphaFold 3-Type Models for Molecular Glue-Induced Ternary Complex Prediction

YDS-GlueFold improves ternary complex prediction by enhancing AlphaFold 3 with guided diffusion, achieving high accuracy on novel protein pairs not seen during training.

MorphoDiff: Cellular Morphology Painting with Diffusion Models

MorphoDiff predicts high-resolution cell morphology changes under different interventions using diffusion models and perturbation embeddings.

Pharmacophore-guided de novo drug design with diffusion bridge

This paper introduces PharmacoBridge, a diffusion bridge-based method for de novo drug design guided by pharmacophore arrangements, generating molecules with high binding affinity to target proteins.

From spiking neuronal networks to interpretable dynamics: a diffusion-approximation framework

The paper introduces a diffusion-approximation framework to model and interpret neuronal spiking dynamics using nonlinear Hawkes processes, showing its accuracy and applicability to human and nonhuman primate data.

Scalable DNA Feature Generation and Transcription Factor Binding Prediction via Deep Surrogate Models

The paper develops a deep surrogate model to generate DNA breathing features efficiently, improving transcription factor binding predictions.

Target-conditioned diffusion generates potent TNFR superfamily antagonists and agonists

This study uses target-conditioned diffusion to design high-affinity TNFR superfamily antagonists and agonists with switchable specificity.

Controllable All-Atom Protein Generation with Latent Diffusion

PLAID generates controllable all-atom protein structures from sequences using a diffusion model trained on sequence data, enabling functional control through Gene Ontology and organism keywords.

SurfDock is a Surface-Informed Diffusion Generative Model for Reliable and Accurate Protein-ligand Complex Prediction

SurfDock improves protein-ligand complex prediction by integrating multiple protein representations into a diffusion model, achieving higher accuracy and physical plausibility than existing methods.

Boltz-1: Democratizing Biomolecular Interaction Modeling

Open-source Boltz-1 predicts biomolecular complex structures comparably to AlphaFold3 and Chai-1 on CASP15 and recent PDB targets; Boltz-steering lets 97% of test-set poses pass physical checks.

RNADiffFold: Generative RNA Secondary Structure Prediction using Discrete Diffusion Models

RNADiffFold predicts RNA secondary structures using discrete diffusion models, showing competitive performance and capturing dynamic structural aspects.

SuperWater: Predicting Water Molecule Positions on Protein Structures by Generative AI

SuperWater predicts water molecule positions on protein structures using a generative AI framework, achieving high accuracy in crystal water coverage and prediction precision.

Generative Models for Prediction of Non-B DNA Structures

The paper evaluates generative models for improving non-B DNA structure prediction through data augmentation, finding diffusion models perform best for Z-DNA and H-DNA while WGAN excels for G-quadruplexes.

Natural Language Prompts Guide the Design of Novel Functional Protein Sequences

This paper introduces BioM3, a framework that uses natural language prompts to design functional proteins, achieving state-of-the-art performance in prediction and generating proteins with native-like structures and function.

Generating Multi-state Conformations of P-type ATPases with a Diffusion Model

The paper introduces a diffusion model to generate diverse conformations of P-type ATPases, demonstrating high accuracy in capturing functional states.