Identifying and Fighting Coronavirus Disease with Machine Learning Tools

The success of the human body in fighting pathogens, including the SARS-CoV-2 that causes coronavirus disease 2019 (COVID-19), relies on lymphocytes (cells of the immune system) and their receptors. Sequencing the adaptive immune receptor repertoire in combination with a machine learning approach allows researchers to classify individuals who carry diseases and investigate immunity. In his new research Prof. Gur Yaari from the Alexander Kofkin Faculty of Engineering, together with a group of researchers and physicians from Israeli universities and hospitals, was the first to train machine learning to identify individuals with COVID-19, as well as determining disease severity. The training was carried out using B cell receptor (BCR) sequencing data, considered particularly challenging due to the high diversity of these receptors thanks to somatic hypermutation (SHM). The alterations in SHM patterns in COVID-19 patients enabled the identification and classification. This groundbreaking research is expected to form a basis to build therapeutic strategies for COVID-19 patients.

Read more >>