Virus Attenuation by Genome-Wide Alterations of Genomic Signatures

Abstract

Despite enormous efforts in biomedical research there is still a shortage of safe and efficacious viral vaccines. To develop a new vaccine is also costly and time consuming. We propose a synthetic biology approach for making viruses less harmful by making many small modifications to their genetic sequences. We establish using computational methods that viruses have pervasive, species-specific genomic signatures, and that alteration of the genomic signature by systematic modifications to a viral genome can be used for synthetic vaccine development purposes. Using a probabilistic model we analyze 28 virus genomes and compute their genomic signatures. We analyze the relationships between viral genomic signatures and those of their hosts, for the purpose of investigating virus-host co-evolution, using machine learning techniques. A synthetic biology concept for altering a viral genome to produce a live vaccine is introduced. The core of this concept is changing the genomic signature of a virus while maintaining its ability to accurately produce all of its proteins. A first step is taken toward a definition of genomic signature distance and a simple greedy algorithmic solution is presented. Our results show that 25 of 28 viruses have pervasive, species-specific genomic signatures, and the remaining 3 have a genomic signature specific down to the taxonomic level of genus, one rank above species. This indicates that viruses have highly optimized genomes, which are the result of evolutionary selection pressures. The method used to examine virus-host co-evolution did not conclusively show a clear adaptation of viral genomes toward their hosts. The greedy algorithm used to alter a viral genome is used to show that it is possible to produce a genetic sequence with a genomic signature farther away from the original than what could be done by random changes, while still maintaining the protein products of the genome. We can therefore present a novel method for live vaccine development that focuses on the issues of safety and efficacy. It addresses in multiple ways the problem of live vaccines reverting back to their original, harmful state. It also ensures that the target organism’s immune system can respond in a manner exactly like it would against the original virus. The future aims are to improve the algorithmic solution and further on to synthesize new viruses, and thereafter perform experiments using real virus particles in living cells.