Illuminating the Transcriptome through the Genome.

David Elliott

doi:10.3390/genes5010235

Abstract

Sequencing the human genome was a huge milestone in genetic research that revealed almost the total DNA sequence required to create a human being. However, in order to function, the DNA genome needs to be expressed as an RNA transcriptome. This article reviews how knowledge of genome sequence information has led to fundamental discoveries in how the transcriptome is processed, with a focus on new system-wide insights into how pre-mRNAs that are encoded by split genes in the genome are rearranged by splicing into functional mRNAs. These advances have been made possible by the development of new post-genome technologies to probe splicing patterns. Transcriptome-wide approaches have characterised a “splicing code” that is embedded within and has a significant role in deciphering the genome, and is deciphered by RNA binding proteins. These analyses have also found that most human genes encode multiple mRNA isoforms, and in some cases proteins, leading in turn to a re-assessment of what exactly a gene is. Analysis of the transcriptome has given insights into how the genome is packaged and transcribed, and is helping to explain important aspects of genome evolution.

Highlights

The completion of the human genome sequence [1,2] brought together key scientific and philosophical questions, including exactly what we are as a species and individuals
The term exon refers to the fact that exon sequences are expressed in the mRNA made from the gene, as opposed to introns which are removed [7]
The availability of genome sequences have allowed system-wide approaches to identify splicing enhancers and silencers that control splicing and led to significant insights into the “splicing code” [29]. These approaches have included machine learning approaches to utilize hundreds of features in pre-mRNAs including motifs bound by RNA binding proteins and RNA secondary structure predictions to predict in vivo splicing decisions [30]

Summary

Introduction

The completion of the human genome sequence [1,2] brought together key scientific and philosophical questions, including exactly what we are as a species and individuals. The availability of genome sequences have allowed system-wide approaches to identify splicing enhancers and silencers that control splicing and led to significant insights into the “splicing code” [29] These approaches have included machine learning approaches to utilize hundreds of features in pre-mRNAs including motifs bound by RNA binding proteins and RNA secondary structure predictions to predict in vivo splicing decisions [30]. Transcriptome-wide CLIP analyses have enabled maps to be drawn of the target sites of RNA binding proteins relative to regulated exons, and these maps can be used to predict mechanisms of splicing control [45]. CLIP tags of the RNA binding protein Tra2β that is needed for splicing inclusion for a regulated cassette exon in the NASP gene are shown in Figure 3 [32] This screenshot was downloaded from the UCSC genome browser [6]. Alternative events are annotated on genome browsers like the UCSC genome browser (e.g., Figure 4) [6]

To What Extent Can Human Complexity Be Ascribed to Alternative Splicing?

Most Splicing Occurs Co-Transcriptionally

Findings

Conclusions