Machine learning based predictive model and systems-level network of host-microbe interactions in post-COVID-19 mucormycosis. (January 2022)

• Record Type: Journal Article
• Authors: Verma, Anukriti; Rathi, Bhawna
• Is Part Of: Microbial pathogenesis. Volume 162(2022)
• Publication Date: 2022-01
• Subjects: Post-COVID-19 mucormycosis -- Machine learning -- Systems biology -- Bioinformatics -- Disease model -- Host-microbe interaction network -- Risk factors -- Drug targets -- Biomarkers -- Prognosis -- Diagnosis -- Treatment; Pathogenic microorganisms -- Periodicals; Pathology, Molecular -- Periodicals; Communicable Diseases -- microbiology -- Periodicals; Communicable Diseases -- parasitology -- Periodicals; Micro-organismes pathogènes -- Périodiques; Pathologie moléculaire -- Périodiques; Electronic journals; 616.9041
• DOI: 10.1016/j.micpath.2021.105324
• Languages: English
• ISSNs: 0882-4010

Abstract

Abstract: Mucormycosis, a rare infection is caused by fungi Mucorales . The affiliation of mucormycosis with Coronavirus disease (COVID-19) is a rising issue of concern in India. There have been numerous case reports of association of rhino-cerebral-orbital, angioinvasive, pulmonary, respiratory and gastrointestinal tract related mucormycosis in patients with history of COVID-19. The immune dysregulation, preposterous use of steroids, interleukin-6-directed therapies and mechanical ventilation in COVID-19 immunocompromised individuals hypothesizes and predisposes to advancement of mucormycosis. The gaps in mode of presentation, disease course, diagnosis and treatment of post-COVID-19 mucormycosis requires critical analysis in order to control its morbidity and incidence and for prevention and management of opportunistic infections in COVID-19 patients. Our study performs machine learning, systems biology and bioinformatics analysis of post-COVID-19 mucormycosis in India incorporating multitudinous techniques. Text mining identifies candidate characteristics of post-COVID-19 mucormycosis cases including city, gender, age, symptoms, clinical parameters, microorganisms and treatment. The characteristics are incorporated in a machine learning based disease model resulting in predictive potentiality of characteristics of post-COVID-19 mucormycosis. The characteristics are used to create a host-microbe interaction disease network comprising of interactions between microorganism, host-microbe proteins, non-specific markers, symptoms and drugs resulting in candidate molecules. R1A (Replicase polyprotein 1a) and RPS6 (Ribosomal Protein S6) are yielded as potential drug target and biomarker respectively via potentiality analysis and expression in patients. The potential risk factors, drug target and biomarker can serve as prognostic, early diagnostic and therapeutic molecules in post-COVID-19 mucormycosis requiring further experimental validation and analysis on post-COVID-19 mucormycosis cases. Highlights: Post-COVID-19 mucormycosis, a rising issue of concern in India. Gaps in mode of presentation, disease course, diagnosis and treatment. Machine learning, systems biology and bioinformatics analysis included. Potential risk factors, drug targets and biomarkers predicted.