Chemical Space Exploration and Machine Learningbased Screening of PDE7A Inhibitors

• DOI: 10.20944/preprints202502.1596.v1
• Publication Date: 2025-02-21T00:15:06Z
• Authors: Yuze Li; Zhe Wang; Shengyao Ma; Xiaowen Tang; Han-Ting Zhang
• Keywords: Chemical space

Background/Objectives: Phosphodiesterase 7 (PDE7), a member of the PDE superfamily, selectively catalyzes hydrolysis cyclic adenosine 3',5'-monophosphate (cAMP), thereby regulating intracellular levels this second messenger and influencing various physiological functions processes. There are two subtypes PDE7, PDE7A PDE7B, which encoded by distinct genes. PDE7 inhibitors have been shown to exert therapeutic potentials in neurological respiratory diseases. However, FDA-approved drugs based on inhibitor still absent, highlighting need for novel compounds advance development. Methods: To address urgent important issue, we conducted comprehensive chemical informatics analysis with potential inhibition using curated database elucidate characteristics highly active inhibitors. Specific substructures that significantly enhance activity inhibitors, including benzenesulfonamido, acylamino, phenoxyl, were identified interpretable machine learning analysis. Subsequently, model employing Random Forest-Morgan pattern was constructed qualitative quantitative prediction Results: As result, 6 inhibitory screened out from SPECS compound library. These exhibited favorable molecular properties potent binding affinities target protein, holding promise as candidates further exploration development Conclusions: Results present study would innovative provide valuable insights future endeavors discovery