COVID-19 Drug Discovery: Generative Approaches

by Jahnvi Sharma

Molecule generation has been one among the fields of drug discovery that are most revolutionized by the implementation of AI over the last decade. VAE is a generator model for enhancing the range of generated information. Autoencoders instruct molecules into a vector that captures properties like bond order, element, and functional group. VAE captures molecules in a latent space. Once captured, variations are made on the first molecule vectors based on desired properties. These will then be decoded back to novel molecules. To optimize the structures, QED, synthetic Accessibility, and LogP regressors were used to improve the latent space variations. In a very totally different approach, science overcame several of the problems with ancient generative models by developing a completely unique advanced deep Q-learning network with fragment-based drug design (ADQN-FBDD). This allowed for the improved exploration of space by grouping SARS-CoV-2 molecules one fragment at a time instead of counting on latent space adjustments. After making connections and rewardful molecules with the foremost druglike connections, a pharmacophore and descriptor filter was used to refine the set. They demonstrated a strong methodology for designing novel, high-binding compounds refined to the structure of SARS-CoV-2 3CLPro.

To design a drug-generative network, the subsequent is necessary: (1) assortment of Druglike Molecules, (2) a illustration of those molecules in silico (i.e., Fingerprints, Tokenizers), (3) a technique of altering molecules to extend diversity, and (4) screening and modification of the altered molecules. Following GAN-related models, Insilico medicine used 3 of its antecedently validated generative chemistry approaches to focus on the main protease, namely, crystal-derived pocked- based generation, homology modeling-based generation, and ligand-based generation. Almost like target-based virtual screening, the main protease has been the main object of interest for scientists for de novo drug discovery. COVID-19.

37 views0 comments

Recent Posts

See All