Leaders in Pharmaceutical Business Intelligence (LPBI) Group

Artificial intelligence predicts the immunogenic landscape of SARS-CoV-2

Reporter: Irina Robu, PhD

Artificial intelligence makes it imaginable for machines to learn from experience, adjust to new inputs and perform human-like tasks. Using the technologies, computer can be trained to achieve specific tasks by processing large amount of data and recognizing patterns. Scientists from NEC OncoImmunity use artificial intelligence to forecast designs for designing universal vaccines for COVID 19, that contain a broad spectrum of T-cell epitopes capable of providing coverage and protection across the global population. To help test their hypothesis, they profiled the entire SARS COV2 proteome across the most frequent 100 HLA-A, HLA-B and HLA-DR alleles in the human population using host infected cell surface antigen and immunogenicity predictors from NEC Immune Profiler suite of tools, and generated comprehensive epitope maps. They use the epitope maps as a starting point for Monte Carlo simulation intended to identify the most significant epitope hotspot in the virus. Then they analyzed the antigen arrangement and immunogenic landscape to recognize a trend where SARS-COV-2 mutations are expected to have minimized potential to be accessible by host-infected cells, and subsequently noticed by the host immune system. A sequence conservation analysis then removed epitope hotspots that occurred in less-conserved regions of the viral proteome.

By merging the antigen arrangement to the infected-host cell surface and immunogenicity estimates of the NEC Immune Profiler with a Monte Carlo and digital twin simulation, the researchers have outlined the entire SARS-CoV-2 proteome and recognized a subset of epitope hotspots that could be used  in a vaccine formulation to provide a wide-ranging coverage across the global population.

By using the database of HLA haplotypes of approximately 22,000 individuals to design  a “digital twin” type simulation to model how efficient various  combinations of hotspots would work in a varied human population. 

SOURCE

https://www.nature.com/articles/s41598-020-78758-5?utm_content=buffer4ebb7