Web Service Development For Visualization of Viral Antigenic Diversity Data

Abstract

Viruses mutate their proteins but not all the time everywhere, so our tool can identify where that virus did not mutate their proteins. This is made worse when your data is simply a set of aligned protein sequences without further quantification over the diversity of between the strains. Thus, we obtained H5N1 HA virus protein dataset from FluDB to acquire virus protein sequences from various strains from different species and locations, exploiting an entropycalculating function and a sequence diversity motif function to measure the diversity of the antigens surrounding a virus membrane, store that processed data into a NoSQL database structure known as pandas and the protein sequence’s source and descriptors into an SQL database, submit those data into a web browser software package known as VADA that uses a Flask python web framework and uses Jinja2 to create an HTML template that extracts the data from the pandas and SQL database. User receives plots, graphs, and paragraphs indicating important regions of diversity and general more-detailed descriptions of a virus protein’s diversity which they can interpret to identify protein sequences that can be vaccine candidates further down the line. The results have matched the previous paper we have used as validation, as the protein dataset we have used has fallen within the parameters of the proteome dataset the previous paper had. Thus, we can conclude that it is possible to create an automated framework for this method of calculating virus diversity.