Jones Susan, Baizan-Edge Amanda, MacFarlane Stuart, Torrance Lesley
Information and Computational Science Group, The James Hutton Institute, Dundee, United Kingdom.
School of Biology, The University of St Andrews, St Andrews, United Kingdom.
Front Plant Sci. 2017 Oct 24;8:1770. doi: 10.3389/fpls.2017.01770. eCollection 2017.
Viruses cause significant yield and quality losses in a wide variety of cultivated crops. Hence, the detection and identification of viruses is a crucial facet of successful crop production and of great significance in terms of world food security. Whilst the adoption of molecular techniques such as RT-PCR has increased the speed and accuracy of viral diagnostics, such techniques only allow the detection of known viruses, i.e., each test is specific to one or a small number of related viruses. Therefore, unknown viruses can be missed and testing can be slow and expensive if molecular tests are unavailable. Methods for simultaneous detection of multiple viruses have been developed, and (NGS) is now a principal focus of this area, as it enables unbiased and hypothesis-free testing of plant samples. The development of NGS protocols capable of detecting multiple known and emergent viruses present in infected material is proving to be a major advance for crops, nuclear stocks or imported plants and germplasm, in which disease symptoms are absent, unspecific or only triggered by multiple viruses. Researchers want to answer the question "how many different viruses are present in this crop plant?" without knowing what they are looking for: RNA-sequencing (RNA-seq) of plant material allows this question to be addressed. As well as needing efficient nucleic acid extraction and enrichment protocols, virus detection using RNA-seq requires fast and robust bioinformatics methods to enable host sequence removal and virus classification. In this review recent studies that use RNA-seq for virus detection in a variety of crop plants are discussed with specific emphasis on the computational methods implemented. The main features of a number of specific bioinformatics workflows developed for virus detection from NGS data are also outlined and possible reasons why these have not yet been widely adopted are discussed. The review concludes by discussing the future directions of this field, including the use of bioinformatics tools for virus detection deployed in analytical environments using cloud computing.
病毒会给多种栽培作物造成严重的产量和品质损失。因此，病毒的检测与鉴定是作物成功生产的关键环节，对世界粮食安全具有重要意义。虽然采用逆转录聚合酶链反应（RT-PCR）等分子技术提高了病毒诊断的速度和准确性，但此类技术仅能检测已知病毒，即每项检测仅针对一种或少数几种相关病毒。所以，如果没有分子检测手段，未知病毒可能会被漏检，检测过程可能会缓慢且昂贵。已开发出同时检测多种病毒的方法，下一代测序（NGS）如今是该领域的主要关注点，因为它能够对植物样本进行无偏差且无需假设的检测。事实证明，开发能够检测感染材料中多种已知和新出现病毒的NGS方案，对于作物、核原种或进口植物及种质来说是一项重大进展，因为在这些作物中，病害症状不明显、不具特异性或仅由多种病毒引发。研究人员想在不知道要寻找何种病毒的情况下回答“这种作物中存在多少种不同病毒？”这一问题：对植物材料进行RNA测序（RNA-seq）就能解决这个问题。除了需要高效的核酸提取和富集方案外，使用RNA-seq进行病毒检测还需要快速且强大的生物信息学方法，以去除宿主序列并进行病毒分类。在这篇综述中，我们讨论了近期在多种作物中使用RNA-seq进行病毒检测的研究，特别强调了所采用的计算方法。还概述了为从NGS数据中检测病毒而开发的一些特定生物信息学工作流程的主要特点，并讨论了这些工作流程尚未被广泛采用的可能原因。综述最后讨论了该领域的未来发展方向，包括在使用云计算的分析环境中部署用于病毒检测的生物信息学工具。
