Molecular Docking Engine
VERA provides comprehensive molecular docking capabilities through AutoDock Vina integration, supporting both CPU and GPU-accelerated versions for drug discovery and virtual screening applications.
AutoDock Vina Capabilities
Comprehensive molecular docking with multiple Vina versions and GPU acceleration.
Multiple Vina versions (1.1.2, 1.2.3-1.2.7) included
GPU acceleration with Vina GPU 2.1 (CUDA)
Automated receptor and ligand preparation
Batch docking for virtual screening
Grid search optimization
Result analysis and visualization
Available Vina Versions
Vina 1.1.2
CPUOriginal AutoDock Vina version
Features:
- •Classic scoring function
- •Well-validated
- •Widely cited
Vina 1.2.3
CPUUpdated scoring function
Features:
- •Improved accuracy
- •Better hydrogen bonding
- •Updated parameters
Vina 1.2.5
CPULatest stable CPU version
Features:
- •Optimized performance
- •Bug fixes
- •Enhanced stability
Vina GPU 2.1
GPU (CUDA)GPU-accelerated for high-throughput
Features:
- •10-50x speedup
- •Batch processing
- •CUDA optimization
Docking Workflow
Standard molecular docking workflow from receptor preparation to result analysis.
Receptor Preparation
Convert protein structure to PDBQT format
Ligand Preparation
Prepare ligands with charges and torsions
Grid Definition
Define search box around binding site
Docking Execution
Run molecular docking simulation
Result Analysis
Analyze binding poses and scores
Visualization
View protein-ligand complexes
Grid Box Parameters
Key parameters for defining the docking search space.
| Parameter | Description | Default Value |
|---|---|---|
| center_x, center_y, center_z | Grid box center coordinates (Å) | Calculated from ligand or pocket |
| size_x, size_y, size_z | Grid box dimensions (Å) | 20 × 20 × 20 |
| exhaustiveness | Search thoroughness | 8 (higher = more thorough) |
| num_modes | Number of binding poses | 9 |
| energy_range | Energy range for poses (kcal/mol) | 3.0 |
CPU vs GPU Performance
Performance comparison for virtual screening of compound libraries.
| Library Size | CPU Time (Vina 1.2.5) | GPU Time (Vina GPU 2.1) | Speedup |
|---|---|---|---|
| 10 compounds | 5 min | 30 sec | 10x |
| 100 compounds | 50 min | 2 min | 25x |
| 1,000 compounds | 8 hours | 15 min | 32x |
| 10,000 compounds | 3 days | 2 hours | 36x |
| 100,000 compounds | 30 days | 20 hours | 36x |
* Performance may vary based on hardware, protein size, and docking parameters.
Best Practices
Recommendations for optimal docking results.
Receptor Preparation
- Remove water molecules (except structural waters)
- Add missing hydrogens at pH 7.4
- Optimize hydrogen bond network
- Check for missing residues/atoms
Grid Box Setup
- Center on known binding site or co-crystal ligand
- Add 10Å padding around reference ligand
- Use exhaustiveness ≥8 for production
- Validate with known binders first
Ligand Preparation
- Generate 3D structures if starting from SMILES
- Optimize geometry before docking
- Calculate Gasteiger charges
- Set rotatable bonds correctly
Result Analysis
- Consider both score and RMSD clustering
- Analyze protein-ligand interactions
- Visual inspection of binding poses
- Compare with known binding modes