|
Dominik Buschmann 1,Anna Haberberger 1 , Benedikt Kirchner 1 , Melanie Spornraft 1 , Irmgard Riedmaier 4,5 , Gustav Schelling 3 , Michael W. Pfaffl 1 1 Department of Animal Physiology and Immunology, TUM School of Life Sciences Weihenstephan, Technical University of Munich, Freising, Germany 2 Institute of Human Genetics, University Hospital, Ludwig-Maximilians-University, Munich, Germany 3 Department of Anesthesiology, University Hospital, Ludwig-Maximilians-University, Munich, Germany 4 Department of Physiology, TUM School of Life Sciences Weihenstephan, Technical University of Munich, Freising, Germany 5 Eurofins Medigenomix Forensik GmbH, Ebersberg, Germany |
Abstract
Small RNA-Seq has revolutionized transcriptomics research in much the same way RT-qPCR did several decades ago. The mas- sively parallel sequencing of short RNA reads has already yielded unprecedented insights in areas such as gene expression pro- filing, clinical diagnostics, and biomarker discovery. Researchers are simultaneously intrigued by the technology’s promises, and challenged with a multitude of hurdles on the way to accurate and meaningful data derived from high-throughput sequencing. Factors impacting experimental outcomes range from wet lab parameters such as sample type and quality, RNA extraction, and library preparation chemistry to processing and analysing mas- sive amounts of data in the post-experimental phase. In addition to the inherent complexity brought about by multiple types of small RNA being studied in various biological scenarios, the choice from a plethora of commercially available sample processing kits, absence of comprehensive guidelines, and suboptimal reporting further complicate the body of literature. Reliable and reproducible research findings can only be realized by rigorous experimental standardization and validation complemented with extensive and transparent reporting of procedures in small RNA-Seq manuscripts. Forming hypotheses based on assumptions developed from flawed data leads to inconsistent findings across the literature and impedes translation of scientific discoveries into much-anticipated clini- cal applications such as molecular diagnostics and development of liquid biopsy-based disease biomarkers. Even though the paral- lels between RT-qPCR and RNA-Seq for nucleic acid quantification are obvious, it could be argued that Next-Generation Sequencing techniques bear even more risk for biases due to the complexity of post-experimental data processing. The widely-adopted MIQE guidelines clearly demonstrate how authoritative guidelines and quality standards translate into improved reporting, better exper- imental setups and, ultimately, valuable applications of research findings. Focusing on the development of biomarker signatures, we herein similarly point out challenges along the small RNA-Seq workflow, report common sources of experimental bias, and call for rigorous quality control and validation in order to generate high-quality, reproducible and meaningful sequencing data.
http://dx.doi.org/10.1016/j.bdq.2017.02.056
| Back to NGS |
|---|
