Emerging deep learning methods for single-cell RNA-seq data analysis

Jie Zheng, Ke Wang
2019 Quantitative Biology  
Deep learning is making major breakthrough in several areas of bioinformatics. Anticipating that this will occur soon for the single-cell RNA-seq data analysis, we review newly published deep learning methods that help tackle computational challenges. Autoencoders are found to be the dominant approach. However, methods based on deep generative models such as generative adversarial networks (GANs) are also emerging in this area. Author summary: Single-cell RNA sequencing (scRNA-seq) and deep
more » ... ning are revolutionizing the fields of biomedicine and artificial intelligence respectively. Due to features of scRNA-seq data (e.g., large sample sizes, high dimensionality), deep learning looks a promising technique for the data analysis. This is the first review at the intersection of deep learning and scRNA-seq technologies. After listing computational challenges, we describe key ideas of representative deep learning methods and compare their strengths and limitations. Autoencoders have been used the most, although generative models are also emerging. We anticipate an explosive development of new methods in this young area.
doi:10.1007/s40484-019-0189-2 fatcat:36nk2crr6rcmtocexv3rjhddhi