DrugLikenZ: An Open-Source Graphical User Interface for Batch Evaluation of Rule-of-Five Compliance

DrugLikenZ: Automated Cheminformatics Screening Tool Overview DrugLikenZ is a Python-based graphical user interface (GUI) designed for the rapid evaluation of chemical libraries against established drug-likeness criteria. The application automates property calculation and rule-of-five compliance checking, providing researchers with an intuitive visual heatmap of their results. By integrating local RDKit calculations with optional PubChem database queries, it offers a high-throughput workflow for preliminary drug discovery and chemical education. Technical Abstract The core concept of this application is to rapidly predict the drug-likeness of chemical compounds by applying Lipinski's Rule of Five (Ro5), a set of physicochemical criteria predictive of oral absorption and permeability. The calculation method involves using RDKit and querying the PubChem database via the compound SMILES string to retrieve or calculate key properties: MW: Molecular Weight HBA: Hydrogen Bond Acceptor count HBD: Hydrogen Bond Donor count logP: Lipophilicity (Partition Coefficient) The application performs a binary evaluation (Pass/Fail) against specific thresholds. A compound is considered drug-like (accepted) if it violates no more than one of these four rules, meaning its total compliance score must be 3 or 4. Key Features Multiple Rulesets: Supports Lipinski's Rule of Five, Miles Congreve et al. Rule of Three (RO3), Muegge Method, and Veber et al. rules. Automated Data Retrieval: Interfaces with the PubChem PUG REST API for compound metadata. Visual Analytics: Generates dynamic, color-coded heatmaps where green represents compliance and red represents violations. Intelligent Formatting: Names of compounds failing the selected criteria are automatically highlighted in red on the axis. Large Dataset Management: Includes a navigation system to handle files with hundreds of compounds. Clean Export: Save accepted candidates directly to a CSV file and export high-resolution heatmaps as PNG images. Scientific Foundation This software implements filtering logic based on: Lipinski, C. A., et al. (1997). Adv. Drug Deliv. Rev. Veber, D. F., et al. (2002). J. Med. Chem. Muegge, I., et al. (2001). J. Med. Chem. Congreve, M., et al. (2003). Drug Discov. Today. Installation & Usage Full documentation, including installation steps and dependency requirements (customtkinter, rdkit, pandas, seaborn, matplotlib, pubchempy), is available in the linked GitHub repository.