(683a) Invited Talk: Ctrl-V: Computational Tracking of Likely Viral Escape Variants By Iterative Integer Optimization
AIChE Annual Meeting
2024
2024 AIChE Annual Meeting
Food, Pharmaceutical & Bioengineering Division
Biosensors for Detecting Viruses and Extracellular Vesicles
Thursday, October 31, 2024 - 12:30pm to 1:10pm
SARS-CoV-2 emerged in China In late 2019 has since had a profound global impact, infecting over 670 million individuals, and resulting in more than 6.8 million fatalities by 2023. Although vaccines have demonstrated effectiveness against the native SARS-CoV-2 strain and its variants, the Omicron variant has emerged as the predominant variant due to critical mutations in its spike protein. These mutations have led to increase in binding affinity to human cell receptor ACE2 (angiotensin converting enzyme 2), allowing for viral entry and antigen replication within the host. In response, the human immune system generates B-cells and T-cells to protect against the virus. Opsonization is a crucial immune mechanism for antigen elimination, involving the binding of antibodies produced by B-cells. The survival of the viral antigen relies on its ability to escape this binding process while maintaining their binding affinity with entry-receptors like ACE2, often achieved through mutations in the spike protein region. A priori knowledge to future infective and/or transmissive strains of a virus will enable humans to be therapeutically prepared with escape-proof antibody formulations. To this end, we introduce a simulation platform CTRL-V (Computational Tracking of Likely Viral escape variants) that employs integer optimization to iteratively model the viral escape process of a viral antigenic protein when confronted with human antibodies. We show a test case demonstration in correctly recovering the infective strains of SARS-CoV-2 starting with the wildtype spike receptor-binding domain against known commercial neutralizing antibodies. Leveraging only the information pertaining to the antibody-antigen complex structure and amino acid interaction preferences in proteins, CTRL-V unveils a putative viral mutational landscape. This information is key to surveillance and antibody design strategies for intervention of viral diseases in humans, and livestock.