Experts publish a new set of consensus principles for classifying the virosphere

The official body responsible for virus classification has published four new principles that bring order to the virus world. This provides a unified framework that allows for the classification of all viruses, something much needed as genomic technologies continue to discover millions of new virus species.

Since the ancient civilizations of Greece and Egypt, humans have attempted to classify life on earth, dividing organisms into related groups in order to understand life and infer relationships.

This classification of life, or taxonomy, took a giant step forward in the 18th century when the Swedish botanist Carl Linnaeus introduced a hierarchical classification system that grouped organisms according to common characteristics. Significantly, he developed a Latin naming system that described each organism by group (genera) and specific name (species). Higher ranks brought related genera together into families, families into orders, and so on through classes, phyla, and kingdoms.

With the publication of his On the Origin of Species, Charles Darwin gave us an understanding of how life forms evolved on Earth and a framework by which organisms can be classified or reclassified based on evolutionary relationships. Based on the Linnaeus template, this evolutionary taxonomy has enabled the construction of a tree of life on which the evolution of each organism can be traced back through more closely related organisms on the same branch to a common ancestor.

With advances in genome sequencing, we can now read the genetic code of any organism; determine how organisms are related to each other; and place them more accurately on increasingly detailed taxonomic trees. This genomic taxonomy has been adopted by microbiologists, allowing them to classify microbes that previously could not be distinguished based on physical characteristics alone.

Genome technology has also revealed just how much genetic diversity there is; High-throughput sequencing has shown that there are millions of microbial species, many more than the tens of thousands known and classified to date. Official bodies, including various interest groups, are beginning to grapple with how taxonomy can accommodate this vast, unknown diversity of organisms that we cannot see and sometimes cannot even cultivate.

The body charged with developing and maintaining the taxonomy of viruses and their names is the International Committee on Taxonomy of Viruses (ICTV). In the pre-genomic age, viruses were classified based on the shape of their particles, the type of nucleic acid in their genome, or other physical characteristics such as the disease they cause or the organism they infect. But we now know from genomic samples from different environments that there are millions of virus species, and they are linked to all walks of life.

In 2016, a consensus statement was published accepting the principle that viruses can be classified based on their genome sequence without the need to culture them or describe their physical properties. While this advance recognizes the great diversity, it does not resolve ongoing debates about how viruses should be classified; should it be based on clinical significance, host, or other biological characteristics? Using Darwin’s principles, viruses do not fit into a universal evolutionary taxonomy because they appear to have evolved from a number of different, independent ancient evolutionary origins and share multiple trees of evolutionary history with little or no overlap.

To address this, the ICTV brought together an expert panel of basic and clinical virologists, bioinformaticians, and evolutionary and structural biologists to develop a community-led consensus on viral taxonomy. The results of their discussions have now been published in the journal PLOS biology and describe four central principles for establishing a universal virus taxonomy.

Overall, ICTV proposes that viral taxonomy can and should be formally based on evolutionary relationships between viruses, using other physical properties where appropriate to inform rank placement.

The first proposed principle is that viruses should be classified primarily based on the best possible reconstruction of their evolutionary history. This currently recognizes at least six virus groups or ranges, each of independent origin and sharing “hallmark” genes; This reflects the way biologists construct trees of life back to the cell’s earliest evolutionary origins.

The second principle recognizes that other properties that naturally group viruses may be of use to researchers studying them; for example, their host area or geographic distribution. For official taxonomic classification, these properties can inform – but not override – evolutionary relatedness in the placement of viruses in the overall taxonomy.

The third principle recognizes that there can be useful ways to group viruses that completely ignore evolutionary relationships. For example, the term arbovirus, derived from viruses born in arthropods, is a useful agricultural classification for viruses transmitted by insect vectors. However, this classification groups many viruses that are not related to each other, and so should not be called a virus taxonomy. In general, a universal evolutionary taxonomy does not preclude alternative classifications that may be more appropriate for medical, veterinary, or agricultural purposes.

These principles allow viruses to be seamlessly classified based on their genome sequence, and are already enabling thousands more to position themselves on the various viral life trees. We are in an age where it is possible to deduce many properties of a virus from its genome sequence, provided that sequence is complete and accurate. This guides the fourth principle, which requires rigorous quality control of genomic sequences to ensure they are appropriate for use in taxonomic mapping.

These principles will help break down decades of disagreement over virus classification while leveraging the genomics revolution to provide a viral taxonomy that truly reflects and embraces the vast diversity in the viral world.

dr Evelien Adriaenssens of the Quadram Institute said: “It was an honor to lead and moderate the discussions on such an important topic as the future of virus taxonomy with such a wonderful and respected group of virologists. The consensus we have reached and the principles we have laid down in this paper will provide us with the framework to include the – sometimes infuriating – diversity of viruses discovered through sequence-based studies along with the viruses that we have known and loved for years.”

Prof Peter Simmonds from the University of Oxford said: “It was a wonderful experience to share a workshop with experts from all over the world on viral diversity, genomics and classification. We have taken some important steps to understand each other’s viewpoints, build consensus, and develop a coherent set of principles that will hopefully guide the broader virological community since its inception in the 1960s. Resolving these conflicts offers a clear strategy to harness the genomics revolution in virology.”

Prof. Murilo Zerbini of the Universidade Federal de Viçosa, Brazil, remarked: “The virus taxonomy has been changed (for the best) by the deluge of metagenome data, and we can now fill in many of the gaps that existed in our knowledge of the viral evolutionary relationships between viruses The participants of the workshop agreed (after some intense debates!) that the inclusion of this data in the taxonomic framework allows for the unification of virus taxonomy based on the four principles outlined in the paper The amazing diversity of the virosphere is finally within reach moved.”