June 2022: Tubiana & Wolfson: Epitopes, deep learning and COVID-19

A new study by Dr. Jérôme Tubiana on Epitopes, deep learning and COVID-19 was published in Nature Methods and featured in a Nature Research Briefing.

June 2022: Tubiana & Wolfson: Epitopes, deep learning and COVID-19

Our body is continuously exposed to protein antigens, and responds by the production of antibodies that bind them. Determining which regions of the protein - termed epitopes - are most likely to be targeted by antibodies is an important question. For instance, the majority of vaccine or infection-induced antibodies against SARS-CoV-2 target the spike protein, and more specifically the region that binds the ACE2 protein - the entry point of SARS-CoV-2 in human cells.

 

In a new study Dr.  Jérôme Tubiana, Edmond J. Safra Postdoc fellow (Wolfson lab, Computer Science) developed a deep learning model named ScanNet that addresses the challenge, based on the protein’s 3D structure. ScanNet was found to be faster and more accurate than previous methods, and faithfully determined the epitopes of the SARS-CoV-2 spike protein. The researchers used ScanNet to compare all strains of the SARS-CoV-2 virus, and found that Omicron was less susceptible to antibodies than previous strains, thus providing a possible explanation for it rapid spread. In the future, predicting antibody epitopes using ScanNet will facilitate monitoring of viral variants, vaccine design and design of non-immunogenic therapeutic proteins.

 

The study was published in Nature Methods and was also featured in a Nature Research Briefing, press releases from Nature portfolio.

 

 

Tel Aviv University makes every effort to respect copyright. If you own copyright to the content contained
here and / or the use of such content is in your opinion infringing, Contact us as soon as possible >>