Machine Intelligence in Single-Cell Data Analysis: Advances and New Challenges.

Jiajia Liu,Jiajia Liu,Xiaobo Zhou,Zhiwei Fan,Weiling Zhao,Zhiwei Fan

doi:10.3389/fgene.2021.655536

Abstract

The rapid development of single-cell technologies allows for dissecting cellular heterogeneity at different omics layers with an unprecedented resolution. In-dep analysis of cellular heterogeneity will boost our understanding of complex biological systems or processes, including cancer, immune system and chronic diseases, thereby providing valuable insights for clinical and translational research. In this review, we will focus on the application of machine learning methods in single-cell multi-omics data analysis. We will start with the pre-processing of single-cell RNA sequencing (scRNA-seq) data, including data imputation, cross-platform batch effect removal, and cell cycle and cell-type identification. Next, we will introduce advanced data analysis tools and methods used for copy number variance estimate, single-cell pseudo-time trajectory analysis, phylogenetic tree inference, cell–cell interaction, regulatory network inference, and integrated analysis of scRNA-seq and spatial transcriptome data. Finally, we will present the latest analyzing challenges, such as multi-omics integration and integrated analysis of scRNA-seq data.

Highlights

The rapid development of single-cell sequencing technologies makes it possible to explore cell heterogeneity of genome, epigenome, and transcriptome, and cell–cell interaction/communication in the context of a specific environment in a tissue
Model-based imputation methods are needed for data imputation to clean the technical noise and correct false expression and dropout events
Using either simulated or real scRNAseq data, our analysis indicated that scIGANs significantly enhanced various downstream analyses compared to existing imputation algorithms

Summary

INTRODUCTION

The rapid development of single-cell sequencing technologies makes it possible to explore cell heterogeneity of genome, epigenome, and transcriptome, and cell–cell interaction/communication in the context of a specific environment in a tissue. Since the general batch effect removal algorithms may not be applied to singlecell sequencing data, some computational methods have been developed to address the challenges of cross-platform/protocol single-cell sequencing data integration (Butler et al, 2018; Kiselev et al, 2018; Barkas et al, 2019; de Kanter et al, 2019; Stuart et al, 2019; Song et al, 2020) These methods extract shared information from individual cells across different datasets, but ignore the differences between datasets. Computational tools have been developed to allocate cells to their cell cycle stages based on their transcriptional profiles

Computational Methods to Predict Cell Cycle Phases

Methods