LiProS: Findable, Accessible, Interoperable, and Reusable Data Simulation Workflow to Predict Accurate Lipophilicity...

Area of Science:

• Computational Chemistry
• Cheminformatics
• Drug Discovery

Background:

• Lipophilicity is a key physicochemical property influencing drug absorption, distribution, and biomolecular interactions.
• Accurate prediction of pH-dependent lipophilicity (log D) is crucial for drug design and materials science.
• Existing methods may not adequately capture the nuances of ionizable compounds.

Purpose of the Study:

• To introduce LiProS, a FAIR (Findable, Accessible, Interoperable, Reusable) workflow for determining pH-dependent lipophilicity profiles.
• To provide researchers with an accessible tool for predicting lipophilicity based on SMILES codes.
• To enable the selection of appropriate lipophilicity formalisms for diverse chemical compounds.

Main Methods:

• Development of the LiProS workflow, accessible via Google Colab.
• Utilizing SMILES codes as input for molecular representation.
• Incorporation of the ion apparent partition coefficient (P_I^app) for enhanced accuracy in pH-dependent predictions.
• Application to the NAPRORE-CR natural products database for analysis of ionizable compounds.

Main Results:

• LiProS efficiently determines pH-dependent lipophilicity profiles from SMILES codes.
• The workflow demonstrated utility in analyzing ionizable compounds.
• LiProS facilitated the identification of suitable lipophilicity formalisms for specific compound sets.

Conclusions:

• LiProS offers a user-friendly and FAIR-compliant solution for lipophilicity prediction.
• The tool enhances the accuracy of pH-dependent lipophilicity modeling, particularly for ionizable compounds.
• LiProS supports data management and sharing principles in scientific research.