Single-cell RNA-seq data analysis using graph autoencoders and graph attention networks

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• High-throughput sequencing has led to a surge in single-cell RNA sequencing (scRNA-seq) data.
• scRNA-seq data is characterized by high dimensionality, high dropout rates, and sparsity.
• Accurate gene imputation and cell clustering are crucial for scRNA-seq data analysis, but traditional methods are inefficient.

Purpose of the Study:

• To develop an advanced computational model for scRNA-seq data analysis.
• To improve gene imputation and cell clustering accuracy for scRNA-seq data.
• To overcome limitations of existing deep learning methods in processing non-Euclidean spatial data.

Main Methods:

• Developed scGAEGAT, a multi-modal model integrating graph autoencoders and graph attention networks.
• Utilized graph neural networks for scRNA-seq data analysis.
• Evaluated gene imputation using cosine similarity, median L1 distance, and root-mean-squared error.
• Assessed cell clustering using adjusted mutual information, normalized mutual information, completeness score, and Silhouette coefficient score.

Main Results:

• scGAEGAT demonstrated promising performance in gene imputation tasks.
• The model achieved high accuracy in cell clustering predictions.
• Experimental results were validated on four scRNA-seq datasets with gold-standard cell labels.
• scGAEGAT outperformed existing methods in both imputation and clustering.

Conclusions:

• scGAEGAT offers an effective solution for scRNA-seq data analysis challenges.
• The model advances gene imputation and cell clustering capabilities.
• This graph neural network-based approach provides a robust framework for complex biological data.