Structure-Based Virtual Screening (SBVS)
Target Structure Preparation 
Compound Library Customization 
Docking & Scoring Biological Assay Validation Given a target structure, how do you find small molecules that modulate its activity? Structure-based virtual screening docks compound libraries against the receptor, ranks poses by scoring functions, and delivers hits for experimental validation—faster and more economically than brute-force synthesis.
Why Structure-Based Virtual Screening Is the Critical Bridge Between Structure and Hit?
High-throughput screening synthesizes thousands of compounds before knowing if any bind. For seed-stage biotechs, that burns runway and material budget. For pharma teams, it delays target validation. SBVS replaces this with computational pre-filtering: receptor structure, compound library, molecular docking, scoring, and hit analysis—delivering a prioritized shortlist before a single flask is opened.
What Sets the Platform Apart
Integrated Structure Determination
Target structures sourced from PDB, Homology Modeling & Threading, or determined in-house via AI-Assisted X-ray Crystallography Services, AI-Enhanced Cryo-Electron Microscopy (Cryo-EM) Services, and AI-Enhanced NMR Spectroscopy Services. Binding site prediction via topology, physicochemical properties, and force field analysis.
Customized Compound Libraries
Public, commercial, and proprietary in-house databases tailored to project needs. Libraries filtered by drug-likeness, diversity, and target-class relevance. Custom library preparation available.
Flexible Docking-to-Assay Pipeline
From docking and scoring to Molecular Dynamics (MD) Simulations, Binding Free Energy Calculation (FEP/TI, MM/PBSA), and biological assay validation. One team, one accountability chain.
The Structure-Based Virtual Screening Suite
Target Preparation & Structure Determination
From Gene to Structure to Binding Site
- PDB Database Search — Retrieval of existing crystal structures with resolution and ligand occupancy assessment.
- Homology Modeling & Threading — AlphaFold Protein Structure Prediction and template-based modeling for targets without experimental structures.
- In-House Structure Determination — Gene-to-Protein Production, AI-Assisted X-ray Crystallography Services, AI-Enhanced Cryo-Electron Microscopy (Cryo-EM) Services, and AI-Enhanced NMR Spectroscopy Services for targets requiring de novo structure solution.
- Binding Site Prediction — Protein surface analysis via topology, physicochemical properties, and force field evaluation.
For virtual biotechs without structural biology infrastructure, we deliver the target structure and binding site analysis as part of the SBVS package. For pharma teams, our Structural Data Processing Services refine existing structures to docking-grade quality.
Compound Database Selection & Customization
Public, Commercial, and Proprietary Libraries
- Public Databases — ZINC, ChEMBL, PubChem, and other open-access collections.
- Commercial Libraries — Enamine, ChemDiv, SPECS, and other vendor catalogs.
- Proprietary In-House Databases — Targeted collections curated for drug-likeness, diversity, and specific target classes.
- Custom Library Preparation — Filtering, protonation, tautomer enumeration, and stereochemistry expansion tailored to project requirements.
Library quality determines hit quality. We select and prepare databases based on your target class, desired chemical diversity, and ADMET Prediction & Modeling constraints.
Molecular Docking & Scoring
Binding Conformation Prediction and Rank Ordering
- Rigid and Flexible Docking — Protein-Ligand Docking (Rigid / Flexible / Induced Fit Docking) for pose prediction and side-chain adaptation.
- Scoring Function Ranking — Physics-based and knowledge-based scoring for hit prioritization.
- Postprocessing of Top-Scoring Hits — Interaction fingerprint analysis, strain energy evaluation, and visual inspection.
Docking predicts how compounds fit; scoring ranks which ones fit best. For Lead Optimization, this provides the structural rationale for analog design.
Platform Instrumentation
| Software / System | Core Capability |
|---|---|
| AutoDock Vina 1.2.0 | Ultra-large library virtual screening with batch-mode docking and flexible side-chain sampling. |
| Schrödinger Glide + Prime | High-precision docking and induced-fit refinement with OPLS4 force field. |
| GNINA 1.0 | CNN-enhanced docking with GPU acceleration for pose prediction and affinity estimation. |
| MOE / ICM-Pro | Pharmacophore-constrained docking and multi-target reverse docking. |
| OpenEye OEDocking + Spruce | Structure preparation, protonation-state assignment, and high-resolution pose prediction. |
| GROMACS 2023 + AMBER 22 | All-atom MD for post-docking pose stability and Binding Free Energy Calculation (FEP/TI, MM/PBSA). |
| NVIDIA A100 GPU Cluster | Parallelized docking and large-scale virtual screening. |
| PyMOL + Maestro | Pose visualization, interaction fingerprint analysis, and pharmacophore alignment inspection. |
Standardized Workflow
Project Workflow
A standardized, milestone-driven execution system. From target biology to validated hits—managed by a single project team, tracked in real time.
01 Target Review & Structure Preparation
- Target biology review and PDB database search.
- Structure quality assessment: resolution, B-factors, missing loops.
- Homology modeling or in-house structure determination if needed.
Deliverable: Prepared target structure + binding site map.
02 Compound Database Selection
- Library selection: public, commercial, proprietary, or custom.
- Drug-likeness and diversity filtering.
- Protonation, tautomer, and stereochemistry preparation.
Deliverable: Customized compound library + quality report.
03 Molecular Docking & Scoring
- Molecular docking execution: rigid, flexible, or induced-fit.
- Scoring function ranking and pose prediction.
- Visual inspection and strain energy evaluation.
Deliverable: Ranked docking poses with confidence scores.
04 Hit Analysis & Refinement
- Top-scoring hit clustering and interaction analysis.
- Molecular Dynamics (MD) Simulations for pose stability (if contracted).
- Binding Free Energy Calculation (FEP/TI, MM/PBSA) rescoring (if contracted).
Deliverable: Refined hit dataset with stability metrics.
05 Report & Handoff
- Comprehensive SBVS report with hit list and binding modes.
- Structural rationale for each prioritized hit.
- Biological assay validation planning.
Deliverable: Final report + data package + transition plan to Hit to Lead or biological assay.
Sample Requirements
| Requirement | Details |
|---|---|
| Target information | Official gene symbol, species, PDB ID (if available), or biological context for structure determination |
| Target structure | PDB ID, AlphaFold model, or request for in-house X-ray/Cryo-EM/NMR structure determination |
| Compound library preference | Public, commercial, proprietary, or custom; specify target class or diversity requirements |
| Project scope | Hit identification, lead optimization, or scaffold-hopping |
| Prior data | Any known actives, SAR, or ADMET flags to guide docking and scoring |
Standard Deliverables
- Prepared target structure with binding site analysis and quality assessment
- Customized compound library with preparation documentation
- Ranked docking poses (top 100–500) with 3D coordinates and interaction fingerprints
- Scoring table with confidence intervals and strain energy metrics
- Pose stability validation via Molecular Dynamics (MD) Simulations (if contracted)
- Electronic data package formatted for biological assay or Hit to Lead handoff
Frequently Asked Questions
Case Study
Case Study: Repurposing 13-cis-Retinoic Acid as a PTP1B Inhibitor via Structure-Based Virtual Screening
Published Evidence:
Navarrete-Mondragón RDC, et al. Virtual and in Vitro Screening Employing a Repurposing Approach Reveal 13-cis-Retinoic Acid is a PTP1B Inhibitor. ChemMedChem. 2024;19:e202400452.
Key Findings:
- AlphaFold-Driven VS: PTP1B1–400 AlphaFold model stabilized by 200 ns MD; 1,075 repurposing candidates filtered by chemotype overlap and docked against catalytic and allosteric sites.
- Consensus Docking: Hierarchical screening with AutoDock Vina, AutoDock 4.2, and GOLD identified 13-cis-retinoic acid as a top hit.
- Experimental Validation: Competitive PTP1B inhibition confirmed (IC50 = 44 μM, Ki = 23.4 μM), with MM/PBSA validating stable catalytic site binding (ΔGbind = −45.11 kcal/mol).
From Screen to Repurposed Hit:
This study demonstrates that SBVS against AlphaFold structures—combined with consensus docking and MD validation—can surface commercially available compounds with novel target engagement. Our platform replicates this workflow, pairing target structure preparation with Molecular Dynamics (MD) Simulations and Hit Biophysical Characterization to confirm computational predictions.
Figure 1. The dose-response curve of 3a / IC50 determination. (Navarrete-Mondragón RDC, et al. 2024)
Reference
- Navarrete-Mondragón RDC, et al. Virtual and in Vitro Screening Employing a Repurposing Approach Reveal 13-cis-Retinoic Acid is a PTP1B Inhibitor. ChemMedChem. 2024;19:e202400452.
Need structure-based virtual screening to advance your hit identification pipeline? Our team can design an SBVS campaign tailored to your target, compound library, and screening goals. Contact our scientific team today to start your project.
