Nuffield Department of Medicine, University of Oxford, Oxford, United Kingdom.
Quadram Institute Bioscience, Norwich Research Park, Norwich, United Kingdom.
PLoS Biol. 2023 Feb 13;21(2):e3001922. doi: 10.1371/journal.pbio.3001922. eCollection 2023 Feb.
A universal taxonomy of viruses is essential for a comprehensive view of the virus world and for communicating the complicated evolutionary relationships among viruses. However, there are major differences in the conceptualisation and approaches to virus classification and nomenclature among virologists, clinicians, agronomists, and other interested parties. Here, we provide recommendations to guide the construction of a coherent and comprehensive virus taxonomy, based on expert scientific consensus. Firstly, assignments of viruses should be congruent with the best attainable reconstruction of their evolutionary histories, i.e., taxa should be monophyletic. This fundamental principle for classification of viruses is currently included in the International Committee on Taxonomy of Viruses (ICTV) code only for the rank of species. Secondly, phenotypic and ecological properties of viruses may inform, but not override, evolutionary relatedness in the placement of ranks. Thirdly, alternative classifications that consider phenotypic attributes, such as being vector-borne (e.g., "arboviruses"), infecting a certain type of host (e.g., "mycoviruses," "bacteriophages") or displaying specific pathogenicity (e.g., "human immunodeficiency viruses"), may serve important clinical and regulatory purposes but often create polyphyletic categories that do not reflect evolutionary relationships. Nevertheless, such classifications ought to be maintained if they serve the needs of specific communities or play a practical clinical or regulatory role. However, they should not be considered or called taxonomies. Finally, while an evolution-based framework enables viruses discovered by metagenomics to be incorporated into the ICTV taxonomy, there are essential requirements for quality control of the sequence data used for these assignments. Combined, these four principles will enable future development and expansion of virus taxonomy as the true evolutionary diversity of viruses becomes apparent.
一种通用的病毒分类法对于全面了解病毒世界以及交流病毒之间复杂的进化关系至关重要。然而,病毒学家、临床医生、农学家和其他相关方在病毒分类和命名概念以及方法上存在重大差异。在这里,我们根据专家的科学共识,提供了一些建议,以指导构建一致和全面的病毒分类法。首先,病毒的分类应与它们进化历史的最佳重建结果一致,即分类单元应为单系的。这一病毒分类的基本原则目前仅在国际病毒分类委员会(ICTV)的分类代码中适用于种的等级。其次,病毒的表型和生态特性可以提供信息,但不应超越进化关系在等级放置中的重要性。第三,替代的分类法可以考虑表型属性,例如媒介传播(例如“虫媒病毒”)、感染特定类型的宿主(例如“真菌病毒”、“噬菌体”)或表现出特定的致病性(例如“人类免疫缺陷病毒”),这些分类可能具有重要的临床和监管用途,但通常会创建不反映进化关系的多系类群。然而,如果这些分类法满足特定社区的需求或发挥实际的临床或监管作用,就应该保留它们。然而,它们不应被视为或称为分类法。最后,尽管基于进化的框架使通过宏基因组学发现的病毒能够纳入 ICTV 分类法,但对于用于这些分类的序列数据的质量控制存在基本要求。这四项原则结合在一起,将使病毒分类法在病毒真正的进化多样性变得明显时得到未来的发展和扩展。