Table 3.

Important studies that used bioinformatics in new drug design and discovery.

Study	Reference
Bioinformatics was used to elucidate the 3D structure of SARS-CoV-2 protease, to predict molecular docking and to simulate drug safety, all of which were essential for developing drugs against COVID-19.	Chukwudozie et al. [85]
Computer Aided Drug Design (CADD) rely upon Structure Based Drug Design (SBDD) and Ligand Based Drug Design (LBDD), both of which need high performance algorithms to model realistic interactions among designed molecules with host and viral ones.	Selvaraj et al. [93]
Putative signaling pathways able to modulate infection reduction, due to decreased viral replication and less symptoms associated with SARS-CoV-2, with help of an application programing interface (API), predicting efficacy of a dozen existing human approved drugs, among which drugs used to treat Systemic Lupus Erythematosus (LES) and Multiple Sclerosis (MS).	Scott et al. [90]
Virtual screening along with in silico predictions and molecular docking analysis provided two possible candidate antiviral drugs for COVID-19 and it was possible to simulate patterns of absorption, distribution, metabolism, and toxicity (ADME) for drugs against SARS-CoV-2.	Ghaebi et al. [94], Hage-Melim et al. [95], Rahman et al. [96] and Alghamdi et al. [97]
Molecular docking algorithms such as GLIDE, AutoDock Vina and SwissDock, focusing on interactions between drug molecules and viral proteins using 3D modeling, provide the opportunity to find existing drugs with high affinity for viral proteins, followed by use of other bioinformatics tools to reduce the number of candidate molecules.	Aronskyy et al. [98] and Chaudhari et al. [99]
In silico docking studies have shown that Remdesivir has multiple potential protease inhibition (PLpro) sites and molecular modeling of all SARS-CoV-2 protein structures using homology modeling suggests several drugs as potential antiviral agents, antibiotics, and muscle relaxants.	Wu et al. [100], Sakr et al. [101], Eweas et al. [102] and Frances-Monerris et al. [103]
Many computational tools were rapidly created in face of the mortal danger the world population was facing during the pandemic's first year and Deep Learning based Ligand Design predicts poly pharmacologic profiles in human proteins and putative viral drug targets. Resulting data were stored in PolypharmDB, allowing for future discovery of multiple candidate drugs for COVID-19.	Waman et al. [92]
Computational approaches have directed clues to identify drug candidates that were previously overlooked, giving rise to a crucial new medical field for global emergencies such as the COVID-19 pandemic.	Kumar et al. [104] and Ma et al. [50]