scGAMNN: Graph Antoencoder-Based Single-Cell RNA Sequencing Data Integration Algorithm Using Mutual Nearest Neighbors

It is critical to correctly assemble high-dimensional single-cell RNA sequencing (scRNA-seq) datasets and downscale them for downstream analysis. However, given the complex relationships between cells, it remains a challenge to simultaneously eliminate batch effects between datasets and maintain the topology between cells within each dataset. Here, we propose scGAMNN, a deep learning model based on graph autoencoder, to simultaneously achieve batch correction and topology-preserving dimensionality reduction. The low-dimensional integrated data obtained by scGAMNN can be used for visualization, clustering and trajectory inference. By comparing it with the other five methods, multiple tasks show that scGAMNN consistently has comparable data integration performance in clustering and trajectory conservation.

Authors

• Bai Zhang
• Hanwen Wu
• Yan Wang
• Chenxu Xuan
• Jie Gao Rutgers University

Keywords

• Multi-layer neural network
• Correlation
• Decoding
• Trajectory
• Measurement
• Dimensionality reduction
• Bioinformatics
• Data Integration
• Single-cell Sequencing
• Single-cell RNA Sequencing
• Single-cell Data
• Mutual Nearest Neighbors
• Deep Learning
• Performance Comparison
• Batch Effects
• Clustering Performance
• scRNA-seq Datasets
• Batch Correction
• Cell Types
• Hematopoietic Stem Cells
• Harmony
• Hidden Layer
• Multiple Datasets
• Latent Space
• Feature Matrix
• Low-dimensional Space
• Integration Of Datasets
• Average Silhouette Width
• Graph Convolutional Network
• Cell Dataset
• Semi-supervised Methods
• Pseudotime
• scRNA-seq Data
• Pseudotime Trajectory
• Canonical Correlation Analysis
• Single-cell Transcriptomics
• graph autoencoder
• mutual nearest neighbor
• scRNA-seq data integration
• Humans
• Algorithms
• Cluster Analysis
• Sequence Analysis, RNA
• Single-Cell Analysis
• Gene Expression Profiling