Demystification of RNAseq Quality Control

Abstract

Next Generation Sequencing (NGS) analysis has become a widely used method for studying the structure of DNA and RNA, but complexity of the procedure leads to obtaining error-prone datasets which need to be cleansed in order to avoid misinterpretation of data.  We address the usage and proper interpretations of characteristic metrics for RNA sequencing (RNAseq) quality control, implemented in and reported by FastQC, and provide a comprehensive guidance for their assessment in the context of total RNAseq quality control of Illumina raw reads. Additionally, we give recommendations how to adequately perform the quality control preprocessing step of raw total RNAseq Illumina reads according to the obtained results of the quality control evaluation step; the aim is to provide the best dataset to downstream analysis, rather than to get better FastQC results. We also tested effects of different preprocessing approaches to the downstream analysis and recommended the most suitable approach.