MOATAI-VIR - an AI algorithm that predicts severe adverse events and molecular features for COVID-19's complications

Following SARS-CoV-2 infection, some COVID-19 patients experience severe adverse events caused by pathogenic host responses. To treat these complications, their underlying etiology must be identified. Thus, a novel AI-based methodology, MOATAI-VIR, which predicts disease-protein-pathway relationships for 22 clinical manifestations attributed to COVID-19 was developed. SARS-CoV-2 interacting human proteins and GWAS identified respiratory failure associated risk genes provide the input from which the mode-of-action (MOA) proteins/pathways of the resulting disease comorbidities are predicted. These comorbidities are then mapped to their clinical manifestations. Three uncharacterized manifestation categories are found: neoplasms, mental and behavioral disorders, and congenital malformations, deformations, and chromosomal abnormalities. The prevalence of neoplasms suggests a possible association between COVID-19 and cancer, whether by shared molecular mechanisms between oncogenesis and viral replication, or perhaps, SARS-CoV-2 is an oncovirus. To assess the molecular basis of each manifestation, the proteins shared across each group of comorbidities were prioritized and subject to global pathway analysis. From these most frequent pathways, the molecular features associated with hallmark COVID-19 phenotypes, such as loss of sense of smell/taste, unusual neurological symptoms, cytokine storm, and blood clots were explored. Results of MOATAI-VIR are available for academic users at: http://pwp.gatech.edu/cssb/MOATAI-VIR/.