SARS-COV-2 has sparked a call to action in the scientific community to stop the pandemic from spreading. As of this writing, there are no authorized vaccines or new antiviral medicines that can be used as a front-line defense. By illuminating uncharted viral pathways, an understanding of COVID-19's pathology could help researchers find effective antivirals. Utilizing computational approaches to find novel candidate medications and vaccines in silica is one way to do this. Over the past ten years, the development of efficient viral therapeutics has been made possible by machine learning-based models that were trained on particular proteins. These models are capable of structurally based prediction of inhibitor candidates given a target biomolecule. If a model is given enough information, it can help in the hunt for a medication or vaccine candidate by spotting trends in the data. In this review, we concentrate on the most recent developments in the use of artificial intelligence in the creation of COVID-19 drugs and vaccines, as well as the potential of intelligent training for the identification of COVID-19 therapies. We highlight numerous COVID-19 molecular targets, whose inhibition may improve patient survival, to simplify applications of deep learning for SARS-COV-2. Additionally, we offer Corona -AI, a dataset of substances, peptides, and epitopes identified in vitro or in silica that may be utilized to train models to identify COVID-19 therapy. Since its first outbreak in 2002, the virus family known as Coronaviridae, which causes symptoms similar to pneumonia, has posed a concern on a global scale. Middle Eastern Respiratory Syndrome (MERS) and Severe Acute Respiratory Disease (SARS), which first appeared in 2002 and 2013, respectively, both produced illnesses characterized by gastrointestinal and pulmonary dysfunction. A third Coronavirus outbreak occurred in 2019, and the virus that caused COVID-19, which has symptoms ranging from the common cold to more serious respiratory failure, has been identified as SARS-COV-2.COVID-19 has spread and infected at least 20 million people despite being classified as a pandemic by the World Health Organization (WHO), with a death toll of over 500,000 at the time of this study. Hospitals are using trial and error methods to find COVID-19 drugs, however because of the inefficiency of lab-based high throughput screening (HTS), virtual screening (VS) has become a preferred technique for finding powerful molecules. In order to suppress the growth and/or activation of a cell, VS for rational drug discovery generally includes computationally targeting a specific biomolecule (such as DNA, protein, RNA, or lipid). Two significant subgroups of this kind of screening are ligand-based drug discovery and design and structure-based drug discovery and design. Given that we have access to both experimentally and computationally determined viral protein structures, VS offers a quick and economical method for locating potential antiviral options. A suitable vaccination against a particular infection may take many years to create, and standard vaccine development approaches have proved expensive. Due in part to the fact that bacterial culturing was no longer necessary for discovering vaccine targets, the science of vaccine design underwent a revolution in the early 1990s with the advent of a genome-based technique known as "Reverse Vaccinology" (RV).
HOW AI IS USED IN THE DISCOVERY OF COVID-19 DRUGS AND VACCINE?
About Author
You must be logged in to post a comment.