Challenges of using RNA-seq in the clinical setting

Abstract

RNA-seq is a mature and well-established method for studying the complexity of the transcriptome in the research setting. As this method moves from the research realm to the clinical context, new opportunities for the development of bioinformatics methods arise. During this talk I will present some of the challenges we have found during our work to release a clinical test for tumor samples using RNA-seq. During the first part of the talk I will focus on fusion detection, how it is affected by the degradation of the sample and how to quantify such effect using Fusion Sense [1]. I will also comment on the opportunities and challenges of annotating and predicting the clinical importance of fusions. During the second part of the talk I will comment on variant calling in RNA-seq and how to account for the effects of library preparation by using RVboost [2]. I will then show some preliminary work of leveraging this method in the context of estimating tumor mutational burden in Formalin-Fixed Paraffin-Embedded (FFPE) samples.