Applications of Surflex-Dock Software

Surflex-Dock is a molecular docking module in SYBYL software, which uses a unique empirical scoring function and a search engine based on molecular similarity to dock ligand molecules to protein binding sites. Surflex-Dock uses the protomol to represent the binding pocket of the protein, and uses probes to probe the surface hydrophobicity, hydrogen bonding and electrostatic properties of the protein pocket to generate a negative image of the protein's active pocket. Surflex-Dock introduces four molecular docking modes: Surflex-Dock Normal, Surflex-Dock Screen, Surflex-Dock Geom, and Surflex-Dock GenomX, which can realize flexible protein docking, restricted docking, DNA-targeted docking and other special docking modes. We use Surflex-Dock software to study the binding mode and interaction between small molecule ligands and macromolecular proteins, and predict or explain the activity of molecules.

Our Services Based on Surflex-Dock Software

The purpose of molecular docking is to find the optimal binding mode of ligand molecules and receptor molecules. Therefore, with the support of Surflex-Dock software, our team strives to solve the most important problems facing molecular docking to find the best binding site and evaluate the binding strength between docked molecules.

• Small Molecule Energetics and Conformations Analysis

Our method relies entirely on structure-generated force fields, is accurate for typical drug-like ligands, and has better coverage of different conformations, but is not suitable for known structural libraries.

• 2D to 3D conversion
• Chirality detection and enumeration
• Protonation
• Conformer generation
• De novo drug design

• Molecular Similarity

Our approach is based on a fundamentally different representation of molecular shape. We take the molecular surface directly comparing coulomb electrostatic fields, and we model hydrogen bonding in a direction-sensitive manner.

• Virtual screening
• Pose prediction
• Multiple ligand alignment

• Docking and Binding Site Analysis

Our method is fully automated in aligning and selecting appropriate binding site variants. Powerful use of fully automatic modes is for virtual screening and pose prediction.

• Biomacromolecule homology modeling
• Large-scale PDB retrieval and processing
• Modeling ligand binding as a real-space refinement of conformational ensembles
• Surface-based binding site alignment using the optimal pocket variant selection
• Posture prediction

• Affinity Prediction

Affinity prediction is an extremely challenging field, especially when the target structure is unknown. Our research employs machine learning (not directly physics-based simulations) to accomplish the following subtasks:

• Multiligand alignment of molecular series including multiple scaffolds
• Merge known binding site information
• Binding site model induction using a multi-instance learning approach
• Predict the binding affinity and binding mode of novel ligands

Our Capabilities

At CD ComputaBio, what we can simulate with the support of Surflex-Dock software includes modeling of small drug molecules, conformational analysis, 3D quantitative structure-activity relationship studies, pharmacophore modeling, virtual screening, and homology of biological macromolecules modeling, active site analysis, database searches,etc.

Features of Surflex-Dock Software

• Very extensive validation on public benchmarks
• Highly accurate non-homologous ligand docking using a hybrid approach using ligand similarity
• Directly applicable to synthetic macrocycles with similar accuracy to non-macrocycles
• Surflex-Dock uses a unique empirical scoring function and a molecular similarity-based search engine to dock ligand molecules to protein binding sites
• Surflex-Dock uses a prototype molecule to represent the binding pocket of a protein