Computational Intelligence Methods for the Diagnosis of COVID-19

Abstract

COVID-19 is a global catastrophe affecting over 200 nations worldwide. The RT-PCR and antibody based detection assay are considered as a standard diagnostic method and are expensive, time-consuming with low sensitivity (40–80%) making it imperative to look for alternatives. COVID-19 patients display severe lung damage which can be spotted in chest X-ray and CT scan which can be analyzed by computational methods for confirmation of the disease. Apart from above techniques, we have discussed battery of computational intelligence tools like docking, molecular dynamics and quantum mechanics (QM), the novel biosensor based diagnostic tool has been developed. The biosensor involves hybridization of SARS-CoV-2 RNA with cDNA to form RNA (viral)–cDNA (probe) hybrid, with intercalation of transition metal Osmium-Ruthenium (II) redox probe that release electrons. The electrons generated by redox metal measured by respective electrodes and detect even miniscule quantities of viral-RNA in the sample. The computational methods support efficient measurement of interaction of redox probe with viral genetic material and intercalation of the Osmium-Ruthenium (II) redox probe between RNA-cDNA hybrid. The computational docking offers proof of concept with better sensitivity, speed and accuracy, by presenting stronger interaction of RNA–DNA hybrid with Osmium (II) redox probe as compared DNA-DNA hybrid in RT-PCR.