ASCOT identifies key regulators of neuronal subtype-specific splicing

Jonathan P Ling,Bo Pang,Lizhi Jiang,Patrick J Leavey,Seth Blackshaw,Anand Venkataraman,Devlina Ghosh,Rone Charles,Clayton P Santiago,Abhinav Nellore,Ben Langmead,Brian S Clark,Christopher Wilks

doi:10.1038/s41467-019-14020-5

Abstract

Public archives of next-generation sequencing data are growing exponentially, but the difficulty of marshaling this data has led to its underutilization by scientists. Here, we present ASCOT, a resource that uses annotation-free methods to rapidly analyze and visualize splice variants across tens of thousands of bulk and single-cell data sets in the public archive. To demonstrate the utility of ASCOT, we identify novel cell type-specific alternative exons across the nervous system and leverage ENCODE and GTEx data sets to study the unique splicing of photoreceptors. We find that PTBP1 knockdown and MSI1 and PCBP2 overexpression are sufficient to activate many photoreceptor-specific exons in HepG2 liver cancer cells. This work demonstrates how large-scale analysis of public RNA-Seq data sets can yield key insights into cell type-specific control of RNA splicing and underscores the importance of considering both annotated and unannotated splicing events.

Highlights

Public archives of next-generation sequencing data are growing exponentially, but the difficulty of marshaling this data has led to its underutilization by scientists
RNA splicing is governed by both cis-regulatory elements and transacting splicing factors (RNA-binding proteins that can act as either splicing enhancers or repressors)
alternative splicing catalog of the transcriptome (ASCOT)’s user interface and associated splicing/expression data sets are openly available at http://ascot.cs.jhu.edu

Summary

Introduction

Public archives of next-generation sequencing data are growing exponentially, but the difficulty of marshaling this data has led to its underutilization by scientists. To demonstrate the utility of ASCOT, we identify novel cell type-specific alternative exons across the nervous system and leverage ENCODE and GTEx data sets to study the unique splicing of photoreceptors. The vertebrate nervous system derives much of its transcriptomic and proteomic diversity from highly specific alternative splicing patterns that are not present elsewhere in the body[34] Many neuronal subtypes, such as rod photoreceptors, exhibit alternative exons that are only detected in that specific cell type[35,36,37,38]. Understanding how photoreceptor-specific splicing patterns emerge may facilitate development of cell-based regenerative strategies for treating photoreceptor dystrophies

Methods

Results

Conclusion