QSAR-ME Profiler 2025: A New Software for QSA(P)R Predictions Supported by Structural Analysis

Area of Science:

• Computational chemistry
• Medicinal chemistry
• Toxicology

Background:

• Quantitative structure-activity relationships (QSARs) are computational methods used to predict the biological activity or properties of chemical compounds.
• Over 20 years, the QSAR research unit at the University of Insubria has developed numerous QSAR models for various endpoints.
• Existing QSAR software can be complex to use and may lack features for evaluating prediction uncertainty and applicability domains.

Purpose of the Study:

• To introduce QSAR-ME Profiler 2025, a new software designed to streamline the application and evaluation of QSAR models.
• To provide enhanced features for quantifying prediction uncertainty and defining applicability domains.
• To support structural similarity analysis, customized model application, and QSAR reporting.

Main Methods:

• Development of the QSAR-ME Profiler 2025 software.
• Integration of existing QSAR models from the University of Insubria.
• Implementation of novel algorithms for uncertainty quantification and applicability domain assessment.
• Inclusion of tools for structural similarity analysis and customized model application.

Main Results:

• QSAR-ME Profiler 2025 offers a user-friendly interface for applying and evaluating QSAR models.
• The software provides transparent outputs in both tabular and graphical formats.
• Innovative features allow for the quantification of prediction uncertainty and the definition of applicability domains.
• The software supports structural similarity analysis, custom model integration, and streamlined QSAR reporting.

Conclusions:

• QSAR-ME Profiler 2025 enhances the usability and reliability of QSAR modeling.
• The software facilitates informed decision-making in drug discovery and chemical safety assessment.
• It represents a significant advancement in the field of computational toxicology and cheminformatics.