Standardizing Next-Generation Sequencing Experiments and Analysis Methods

To the Editor: The past few years have seen the emergence of new strategies for high-throughput DNA sequencing that have invigorated life science research. Because of these technological advances, the available sequencing data continue to expand. Additionally, cost reductions have made access to sequences of entire genomes easier for established laboratories and even easier for small research groups. A lack of consensus exists, however, on how best to design and analyze next-generation sequencing (NGS) studies and how to compare these data to previous and future work. Comparing different sequencing platforms and analysis methods, and thus their interpretations, is becoming more complicated. The main purpose of such studies is somehow getting lost in the process of experimental design as biologists confront new territory with NGS work, which requires handling more complex and powerful statistical methods and produces large data sets. Every new tool or method is being compared with existing methods, and investigators are using statistical arguments to propose that the new methods are better. Kiezun et al. (1) have recently highlighted current discrepancies and technical issues in the exome-sequencing approach and in whole-genome sequencing that could bias the interpretation and calling of variants. This situation calls for attention to establish a positive standard that can minimize discrepancies and increase the dependability of all sequencing studies. Some important questions include: Does the assembled and realigned genome represent the correct sequence of the sample? Are the observed effects real, or are they due to differences in analysis methods? Currently, there is no gold standard for high-throughput sequencing experiments and analysis techniques, but now would be an excellent time to develop one. The development of a sequencing gold standard would allow methods to be compared more easily and allow investigators to better understand the interpretations and nuances that derive from each method.