LapDDPM: A Conditional Graph Diffusion Model for scRNA-seq Generation with Spectral Adversarial Perturbations

2025/06/16 by Lorenzo Bini, Bini, Lorenzo, Stéphane Marchand‐Maillet +1
Biochemistry, Genetics and Molecular Biology · #Advanced biosensing and bioanalysis techniques #Artificial Intelligence (cs.AI) #Biomolecules (q-bio.BM) #Cell Behavior (q-bio.CB) #FOS: Biological sciences #FOS: Computer and information sciences #Genomics (q-bio.GN) #Machine Learning (cs.LG) #Molecular Biology Techniques and Applications #RNA Interference and Gene Delivery

paper · pdf · doi:10.48550/arxiv.2506.13344

openalex publication_date 2025/06/16 · openalex created_date 2025/10/13 · openalex updated_date 2026/07/28

Abstract

Generating high-fidelity and biologically plausible synthetic single-cell RNA sequencing (scRNA-seq) data, especially with conditional control, is challenging due to its high dimensionality, sparsity, and complex biological variations. Existing generative models often struggle to capture these unique characteristics and ensure robustness to structural noise in cellular networks. We introduce LapDDPM, a novel conditional Graph Diffusion Probabilistic Model for robust and high-fidelity scRNA-seq generation. LapDDPM uniquely integrates graph-based representations with a score-based diffusion model, enhanced by a novel spectral adversarial perturbation mechanism on graph edge weights. Our contributions are threefold: we leverage Laplacian Positional Encodings (LPEs) to enrich the latent space with crucial cellular relationship information; we develop a conditional score-based diffusion model for effective learning and generation from complex scRNA-seq distributions; and we employ a unique spectral adversarial training scheme on graph edge weights, boosting robustness against structural variations. Extensive experiments on diverse scRNA-seq datasets demonstrate LapDDPM's superior performance, achieving high fidelity and generating biologically-plausible, cell-type-specific samples. LapDDPM sets a new benchmark for conditional scRNA-seq data generation, offering a robust tool for various downstream biological applications.

Citations

• Adversarial Training for Graph Neural Networks: Pitfalls, Solutions, and New Directions
• Spectral Augmentations for Graph Contrastive Learning
• Equivariant Diffusion for Molecule Generation in 3D
• High-Resolution Image Synthesis with Latent Diffusion Models
• High-Resolution Image Synthesis with Latent Diffusion Models
• Diffusion Models Beat GANs on Image Synthesis
• Score-Based Generative Modeling through Stochastic Differential Equations
• Denoising Diffusion Probabilistic Models
• Graph Random Neural Network for Semi-Supervised Learning on Graphs
• Single-cell transcriptomics of 20 mouse organs creates a Tabula Muris
• RNA velocity of single cells
• Adversarial Attack on Graph Structured Data
• Density Estimation Using Real NVP
• Deep Unsupervised Learning using Nonequilibrium Thermodynamics
• Explaining and Harnessing Adversarial Examples
• NICE: Non-linear Independent Components Estimation

Related

• scDD: Latent Codes Based scRNA-seq Dataset Distillation with Foundation Model Knowledge
• scDiffusion: conditional generation of high-quality single-cell data using diffusion model
• scRDiT: Generating single-cell RNA-seq data by diffusion transformers and accelerating sampling
• Single-Cell RNA-seq Synthesis with Latent Diffusion Model
• Shared Differential Clustering across Single-cell RNA Sequencing Datasets with the Hierarchical Dirichlet Process