Table 4.
AI outcomes in SARS-CoV-2.
| AI Model | Results | Study |
|---|---|---|
| Diagnosis | ||
| TopNetmAb model: Comprehensive topology-based AI. | Predict the binding free energy changes of S and ACE2/antibody complexes induced by mutations on the S RBD, of the Omicron variant. | Chen et al., (2022) [113] |
| DL method (3D-DL framework) for DNA sequence classification using CNN. | SARS-CoV-2 viral genomic sequencing. Viral evaluation accuracy > 99%. |
Lopez-Rincon et al., (2021) [114] |
| Drug discovery | ||
| DeepH-DTA: A squeezed-excited dense convolutional network for learning hidden representations within amino acid sequences. | Predict the affinity scores of drugs against SARS-CoV-2 amino acid sequences. | Abdel-Basset et al., (2020) [115] |
| Estimated drug–target interactions. A list of antiviral drugs was identified. | Molecule transformer–drug target interaction (MT-DTI). | Beck et al., (2020) [116] |
| AI-based generative network complex | Generate 15 potential drugs. | Gao et al., (2020) [117] |
| ChemAI; a deep neural network protocol on three drug discovery databases. | Generate 30,000 small compounds that are SARS-CoV-2 inhibitors. | Hofmarcher et al., (2020) [118] |
| ADQN-FBDD: An advanced deep Q-learning network with the fragment-based drug design (a model-free reinforcement learning algorithm). | Generate 47 lead compounds, targeting the SARS-CoV2 3C-like main protease. | Tang et al., (2020) [119] |
| Dense fully convolutional neural network (DFCNN). A list of chemical ligands and peptide drugs was provided. |
Used four chemical compound and tripeptide databases to identify potential drugs for COVID-19. | Zhang et al., (2020) [109] |
| Generative DL. An AI-based drug discovery pipeline. | Generate inhibitors for the SARS-CoV-2 3CLpro. | Zhavoronkov et al., (2020) [120] |
| Vaccine development | ||
| Bioinformatic tools and databases | Epitope vaccines were designed by using protein E as an antigenic site. | Abdelmageed et al., (2020) [121] |
| Computational methodology | Identify several epitopes in SARS-CoV-2 for the development of potential vaccines. S protein was identified as an immunogenic and effective vaccine candidate. |
Fast et al., (2020) [122] |
| ML and reverse vaccinology | A cocktail vaccine with structural and non-structural proteins in which would accelerate efficient complementary immune responses. | Ong et al., (2020) [123] |
| Integrated bioinformatics pipeline that merges the prediction power of different software (in silico pipeline). | Predict the cross-reactivity of pre-existing vaccination interventions against SARS-CoV-2. | Russo et al., (2021) [124] |
| Immune informatics, reverse vaccinology, and molecular docking analysis. | Three epitope-based subunit vaccines were designated. Only one was reported as the best vaccine. | Sarkar et al., (2020) [125] |
| In silico approach. A molecular docking analysis. | A multi-epitopic vaccine candidate targeting the non-mutational immunogenic regions in envelope protein and surface glycoprotein of SARS-CoV-2. | Susithra Priyadarshni et al., (2021) [126] |
3CLpro, 3C-like protease; AI, artificial intelligence; CNN, convolutional neural network; COVID-19, coronavirus disease 2019; DL, deep learning; HGAT, heterogeneous graph attention; ML, machine learning; RBD, receptor-binding domain; SARS-CoV-2, severe acute respiratory syndrome coronavirus 2.