BioPharmics

BioPharmics offers a comprehensive platform for 3D molecular design and visualization, enabling users to understand structure-activity and structure-property relationships through advanced computational modeling. It addresses challenges in predicting molecular conformations and binding affinities, supporting drug discovery efforts from small molecules to large peptidic macrocycles.

From 2D to 3D

Transitioning from 2D molecular images to 3D models involves generating low-energy conformational ensembles while considering chirality and protonation at physiological pH. The platform utilizes the ForceGen methodology for efficient and accurate conformer generation, even for complex macrocycles.

Structure-based Drug Design

Surflex-Dock enhances docking workflows with automated procedures, improving the reliability of docking models and virtual screening. It accounts for protein conformation variability, providing a realistic picture of molecular docking. The combination of eSim and Surflex-Dock offers superior pose prediction, aiding in the design of high-affinity compounds.

No Target Structure? No Problem

For projects lacking target structural information, 3D ligand-based drug design methods like eSim create molecular surface representations based on shape, electrostatic fields, and hydrogen-bond preferences. This approach facilitates the discovery and optimization of novel active compounds, even with multiple active ligands.

Accurate Affinity Prediction

The QuanSA method predicts binding affinity with accuracy comparable to leading structure-based methods, offering greater transferability to novel chemistries. It builds accurate affinity models without protein structure information, using machine learning models that are scaffold-independent and provide prediction confidence metrics.

Macrocycle Modelling

Macrocycles, despite their complexity, can be effectively modeled using BioPharmics. The platform supports macrocyclic lead optimization by accurately predicting conformations and properties, applying biophysical constraints, and utilizing both structure-based and ligand-based machine learning approaches.

Linking 2D and 3D SAR

Integrating 3D structure- and ligand-based design with predictive modeling and SAR analysis provides a comprehensive view of compound activity and quality. The StarDrop’s SeeSAR modules facilitate visualization, binding affinity assessment, pose generation, and interactive 3D design, compatible with major docking software platforms.