Chemical space exploration with Molpher: Generating and assessing a glucocorticoid receptor ligand library

• DOI: 10.26434/chemrxiv-2023-4jm5t-v2
• Publication Date: 2024-02-06T11:16:47Z
• Authors: Maria Isabel Agea; Ivan Čmelo; Wim Dehaen; Ya Chen; Johannes Kirchmair; David Sedlák; Petr Bartůněk; Martin Šícho; Daniel Svozil
• Keywords: Chemical space

Computational exploration of chemical space is crucial in modern cheminformatics research for accelerating the discovery new biologically active compounds. In this study, we present a detailed analysis library potential glucocorticoid receptor (GR) ligands generated by molecular generator, Molpher. To generate targeted GR and to construct classification models, structures from ChEMBL database as well internal IMG library, that was experimentally screened biological activity primary luciferase reporter cell assay, were utilized. The composition ligand compared with reference randomly samples space. A random forest model used determine ligands, incorporating its applicability domain using conformal prediction. It demonstrated significantly enriched library. Furthermore, prospective Molpher successfully designed compounds which subsequently confirmed be on GR. collection 34 also identified. Moreover, an important contribution study establishment comprehensive workflow evaluation computationally particularly those against targets are challenging dock.