AI Platform Identifies 68 Genes Associated with High Risk of Severe COVID-19

By HospiMedica International staff writers
Posted on 02 Jul 2020

An artificial intelligence (AI) platform has found 68 genes associated with risk of developing severe COVID-19 after analysis of the genomes of 929 patients who had a severe response to SARS-CoV-2.

The unique AI platform developed by PrecisionLife Ltd. (Oxfordshire, UK) could help identify patients who are at greatest risk of developing the most severe forms of COVID-19, prior to an effective and widely available vaccine. Established Genome Wide Association Studies (GWAS) approaches have been limited by the heterogenous nature of COVID-19, making it difficult to clearly explain the wide range of symptoms and impacts of predisposing comorbidities associated with the disease. PrecisionLife has overcome this barrier by evaluating combinations of genetic features, which is not possible with existing GWAS approaches.

Image: Disease architecture of the severe COVID-19 patient population generated by the PrecisionLife platform. Each circle represents a disease-associated SNP genotype, edges represent co-association in patients, and colours represent distinct patient sub-populations (Photo courtesy of PrecisionLife Ltd.)

Using a combinatorial (high-order epistasis) analysis approach, PrecisionLife identified 68 protein-coding genes that were highly associated with severe COVID-19, nine of which have been previously linked to differential response to SARS-CoV-2. These 68 genes include several druggable protein targets and pathways, nine of which are targeted by drugs that have reached at least Phase I clinical trials.

“This is our second study specifically looking at host genomics to identify those most at risk and find opportunities to treat patients with later stage severe disease where host immune processes begin to run out of control” said Dr Steve Gardner, CEO of PrecisionLife. “We know that systemic remodelling, leaky vasculature and micro-clotting can cause severe issues in COVID-19 patients in lungs as well as other organs. We are encouraged that several of the genes identified relate to lipid programming, beta-catenin and protein kinase C signalling whose processes converge in a central pathway involved in plasma membrane repair, clotting and wound healing. This pathway is largely driven by calcium ion activation, which is a known serum biomarker associated with severe COVID-19 and ARDS. We intend to perform further analyses to investigate this hypothesis. We are also seeking to further validate 12 genes associated with dysfunctional immune response that are the first hint at a potential genetic signature for enhanced risk of flipping into a severe disease state.”

Related Links:
PrecisionLife Ltd.