Goals and approaches for each processing step for single-cell RNA sequencing data

Single-cell RNA sequencing (scRNA-seq) has enabled researchers to study gene expression at the cellular level. However, due to the extremely low levels of transcripts in a single cell and technical losses during reverse transcription, gene expression at a single-cell resolution is usually noisy and highly dimensional; thus, statistical analyses of single-cell data are a challenge. Although many scRNA-seq data analysis tools are currently available, a gold standard pipeline is not available for all datasets. Therefore, a general understanding of bioinformatics and associated computational issues would facilitate the selection of appropriate tools for a given set of data. In this review, we provide an overview of the goals and most popular computational analysis tools for the quality control, normalization, imputation, feature selection and dimension reduction of scRNA-seq data.

Automated summary

Single-cell RNA sequencing has revolutionized the study of gene expression at a cellular level, but the noise and dimensionality of the data pose challenges for statistical analysis. While there are many tools available for scRNA-seq data analysis, a universal gold standard pipeline is still lacking. Understanding bioinformatics and computational issues can help in selecting the appropriate tools for data analysis.