UC Berkeley

RNA sequencing (RNA-Seq) is a powerful technique to assess gene expression and to characterize RNA transcripts, as it allows for high-throughput study of RNA at a nucleotide resolution. It can provide transcriptome-wide information on alternative splicing, novel transcripts, tissue-specific and allele-specific expression, and RNA- DNA differences. While high-throughput sequencing has revolutionized RNA research, there are several challenges associated with its application. The limiting factors in the utility of any RNA-Seq experiment are two-fold: (1) at the experimental stage, correctly isolating and preparing the sample so as to capture all transcripts of interest, and (2) at the data analysis stage, utilizing the right computational methods to accurately assign sequencing reads to the correct transcripts. Non-canonical isoforms can easily be missed by standard methods at either stage. As RNA exists in a multitude of forms corresponding to its diverse functions, with variations in sequence, length, shape, and number, it is important to develop sequencing methods that are capable of capturing this diversity.

Herein, we describe methods for performing RNA sequencing experiments and data analysis, illustrate their utility in experimental studies using Drosophila melanogaster, and present novel methods for the detection, quantification, and characterization of two non-canonical transcript isoforms: the fusion transcript and circular RNA.