Grant: Discovery Grants Program - Individual

Viruses are everywhere. They have significant impacts on the growth and survival of their host populations, and the flow of resources through the living world. In essence, a virus is a genome made of DNA or RNA that is packaged into a shell of proteins. Because viruses are relatively simple in their biology, we can learn a lot about a virus by examining its genome. RNA viruses have very short genomes, often shorter than a single human gene. A virus has to compress into its genome all the information needed to take over a cell, make copies of itself, and avoid the immune system. RNA viruses solve this problem by minimizing unused parts of the virus genome, and reusing other parts with overlapping genes. Despite these constraints, genomes of related virus species are surprisingly diverse in which genes they contain and how they are arranged. We will develop a method to reconstruct how these genomes have evolved at the level of genes. Building on our previous work, we will use a network to represent the diverse gene composition of genomes in the same virus family. This model will enable us to understand how new genes emerge in virus genomes, and how newly discovered viruses are related to known genomes. The genome of a virus encodes several different proteins. Some of these proteins are located on the surface of the virus, where they enable the virus to bind with and enter a new host cell. This also exposes the proteins to the immune system outside the cell, such as antibodies. We hypothesize that these exposed proteins are under stronger selection than virus proteins that function within the infected cell. To test this, we will generate an "evolutionary fingerprint" of each gene that represents the imprint of selection on the genetic differences that have accumulated between different copies. We will adapt an image recognition technique from artificial intelligence to determine whether surface-exposed proteins have different fingerprints than other virus proteins. This work will shed light on how virus evolution is fundamentally shaped by the host environment. RNA viruses also evolve much faster than the organisms they infect. The enzyme they use to copy their genomes has a high rate of making mistakes (mutations). In fact, many RNA viruses are thought to exist at the brink of an "error catastrophe", beyond which the virus is no longer able to survive from accumulating too many mutations in its genome. However, this model makes some unrealistic assumptions about the combined effects of mutations. Although more realistic models have been developed, they are difficult to fit to real data. Our final objective is to develop a realistic model of mutation accumulation that is easier to fit to experimental data. This model will improve our understanding of the impact of mutation accumulation on the survival of virus populations, which may have selected for antiviral proteins in host species that induce mutations in virus genomes.