In Silico Modeling and Structural Analysis of Soluble Epoxide Hydrolase Inhibitors for Enhanced Therapeutic Design

Area of Science:

• Medicinal Chemistry
• Computational Chemistry
• Pharmacology

Background:

• Human soluble epoxide hydrolase (sEH) is crucial for metabolizing epoxyeicosatrienoic acids.
• sEH inhibitors show therapeutic potential for various diseases.
• Understanding structural requirements for sEH inhibition is vital for drug development.

Purpose of the Study:

• To develop predictive and validated in silico models for soluble epoxide hydrolase (sEH) inhibitors.
• To gain insights into the structural features that enhance inhibitory potential against sEH.
• To provide guidelines for the rational design of novel sEH inhibitors.

Main Methods:

• Utilized diverse in silico modeling approaches, including 2D-QSAR, Transformer-CNN with Layer-wise Relevance Propagation (LRP), and 3D-QSAR.
• Calculated molecular descriptors using multiple tools and employed feature selection strategies.
• Validated QSAR models using molecular dynamics (MD) simulations to analyze receptor-ligand interactions.

Main Results:

• A statistically significant 2D-QSAR model identified topological characteristics, 2D pharmacophore features, and physicochemical properties as critical for inhibitory potential.
• Transformer-CNN models provided structural interpretations of sEH inhibition via LRP.
• 3D-QSAR analysis offered further insights into structural requirements for potent sEH inhibition.
• MD simulations corroborated QSAR predictions by revealing key receptor-ligand interactions.

Conclusions:

• The developed in silico models effectively predict the inhibitory potential of sEH inhibitors.
• Structural insights derived from QSAR and MD simulations are crucial for rational drug design.
• This study provides a framework for designing novel sEH inhibitors with enhanced therapeutic efficacy using open-access tools.