BRIE2: computational identification of splicing phenotypes from single-cell transcriptomic experiments

Guido Sanguinetti,Yuanhua Huang

doi:10.1186/s13059-021-02461-5

Abstract

RNA splicing is an important driver of heterogeneity in single cells through the expression of alternative transcripts and as a determinant of transcriptional kinetics. However, the intrinsic coverage limitations of scRNA-seq technologies make it challenging to associate specific splicing events to cell-level phenotypes. BRIE2 is a scalable computational method that resolves these issues by regressing single-cell transcriptomic data against cell-level features. We show that BRIE2 effectively identifies differential disease-associated alternative splicing events and allows a principled selection of genes that capture heterogeneity in transcriptional kinetics and improve RNA velocity analyses, enabling the identification of splicing phenotypes associated with biological changes.

Highlights

Single-cell RNA-sequencing has rapidly become the key technology to disentangle transcriptional heterogeneity in cell populations
This enables us to statistically associate cell-level features with PSI-values associated with specific splicing events, defining quantitatively splicing as a single-cell level intermediate phenotype, but it considerably increases the complexity of the model
Applying BRIE2 to detect differential momentum genes (DMG) by using the stimulation time as testing covariate, we found 421 DMGs significantly associated with time (ELBO_gain>5; Additional file 1: Figure S19S20), with 201 genes overlapped with the top 2000 highly variable genes selected by scVelo

Summary

Introduction

Single-cell RNA-sequencing (scRNA-seq) has rapidly become the key technology to disentangle transcriptional heterogeneity in cell populations. Over the last 5 years, scRNA-seq has been successfully applied both to identify discrete cell states or subpopulations in normal or diseased tissues, e.g. ScRNA-seq has further been applied to multi-sample designs with different donors, tissues, diseases or treatments. These experiments enable the discovery of cell type specific marker genes [5] or key pathways that are associated with the meta labels [2]. Beyond gene-level information, RNA processing within a gene holds rich information for both categorical cell states and continuous cell differentiation. A key RNA processing step is splicing, where a precursor mRNA (pre-mRNA or unspliced RNA) is spliced by removing intronic, non-coding regions, resulting in mature mRNA

Objectives

Methods

Results

Conclusion