ScGNN is a novel graph neural network framework for single-cell RNA-Seq analyses

Juexin Wang,Hongjun Fu,Jianting Gong,Cankun Wang,Dong Xu,Anjun Ma,Yuexu Jiang,Yuzhou Chang,Qin Ma,Ren Qi

doi:10.1038/s41467-021-22197-x

Juexin Wang, Hongjun Fu + Show 8 more

Open Access

https://doi.org/10.1038/s41467-021-22197-x

Copy DOI

Abstract

Single-cell RNA-sequencing (scRNA-Seq) is widely used to reveal the heterogeneity and dynamics of tissues, organisms, and complex diseases, but its analyses still suffer from multiple grand challenges, including the sequencing sparsity and complex differential patterns in gene expression. We introduce the scGNN (single-cell graph neural network) to provide a hypothesis-free deep learning framework for scRNA-Seq analyses. This framework formulates and aggregates cell–cell relationships with graph neural networks and models heterogeneous gene expression patterns using a left-truncated mixture Gaussian model. scGNN integrates three iterative multi-modal autoencoders and outperforms existing tools for gene imputation and cell clustering on four benchmark scRNA-Seq datasets. In an Alzheimer’s disease study with 13,214 single nuclei from postmortem brain tissues, scGNN successfully illustrated disease-related neural development and the differential mechanism. scGNN provides an effective representation of gene expression and cell–cell relationships. It is also a powerful framework that can be applied to general scRNA-Seq analyses.

Highlights

Single-cell RNA-sequencing is widely used to reveal the heterogeneity and dynamics of tissues, organisms, and complex diseases, but its analyses still suffer from multiple grand challenges, including the sequencing sparsity and complex differential patterns in gene expression
We introduce a multi-modal framework scGNN for modeling heterogeneous cell–cell relationships and their underlying complex gene expression patterns from scRNA-Seq. scGNN trains low-dimensional feature vectors to represent relationships among cells through topological abstraction based on both gene expression and transcriptional regulation information
The left-truncated mixture Gaussian (LTMG) model[14,28] is adopted to the top 2,000 variable genes to quantify gene regulatory signals encoded among diverse cell states in scRNA-Seq data

Summary

Introduction

Single-cell RNA-sequencing (scRNA-Seq) is widely used to reveal the heterogeneity and dynamics of tissues, organisms, and complex diseases, but its analyses still suffer from multiple grand challenges, including the sequencing sparsity and complex differential patterns in gene expression. We introduce the scGNN (single-cell graph neural network) to provide a hypothesis-free deep learning framework for scRNA-Seq analyses This framework formulates and aggregates cell–cell relationships with graph neural networks and models heterogeneous gene expression patterns using a left-truncated mixture Gaussian model. We introduce a multi-modal framework scGNN (single-cell graph neural network) for modeling heterogeneous cell–cell relationships and their underlying complex gene expression patterns from scRNA-Seq. scGNN trains low-dimensional feature vectors (i.e., embedding) to represent relationships among cells through topological abstraction based on both gene expression and transcriptional regulation information. To further demonstrate the power of scGNN in complex disease studies, we applied it to an Alzheimer’s disease (AD) data set containing 13,214 single nuclei, which elucidated its application power on cell-type identification and recovering gene expression values[27] We claim that such a GNN-based framework is powerful and flexible enough to have great potential in integrating scMultiOmics data

Methods

Results

Conclusion