Discovery of Novel GABAAR Allosteric Modulators Through Reinforcement Learning

Abstract

Background: As not all target proteins can be easily screened in vitro, advanced virtual screening is becoming critical. Objective: In this study we demonstrate the application of reinforcement learning guided virtual screening for γaminobutyric acid A receptor (GABAAR) modulating peptides. Method: Structure-based virtual screening was performed on a receptor homology model. Screened molecules deemed to be novel were synthesized and analyzed using patch-clamp analysis. Results: 13 molecules were synthesized and 11 showed positive allosteric modulation, with two showing 50% activation at the low micromolar range. Conclusion: Reinforcement learning guided virtual screening is a viable method for the discovery of novel molecules that modulate a difficult to screen transmembrane receptor.

Synthetic corticotropins and the GABA‐receptor system: Direct and delayed effects

“…Some of these drugs act on proteins involved in synthesis, degradation, and membrane transport of GABA, other—target the GABA receptors functional activity as allosteric modulators (Olsen, 2018). It is known a lot of peptides that are involved in the allosteric modulation of GABA receptors (Farzampour et al, 2015; Michaeli et al, 2020; Tonon et al, 2020). The modulatory effect of such peptides may change in the presence of another allosteric modulator (Kasian et al, 2017) or after exposure to stress (Chigr et al, 2001).…”

Discovery of novel spike/ACE2 inhibitory macrocycles using in silico reinforcement learning

“…To further enhance this process, the concept of risk/return efficiency was translated from the financial world, making the exploitation of existing docked structures for additional sequences more efficient (Lerner et al, 2017). This screening approach has yielded numerous novel polymers; including GABA-A receptor allosteric modulators (Michaeli et al, 2020), bispecific MD2/CD14 macrocycle agonists of Toll-Like Receptor 4 (TLR4) (Michaeli et al, 2018) and a rescue peptide for a destabilizing Glycogen Branching Enzyme (GBE1) mutation (Froese et al, 2015). The latter case showed another advantage for in silico screening; as protein rescue assays had to be conducted in vivo, on primary patient cells, barring the option for massive physical screening (Froese et al, 2015).…”