How to Use GNINA Online
Try GNINA
Commercially Available Online Web Server
GNINA: Molecular Docking with Deep Learning
Scientists have developed GNINA (pronounced "nee-na"), an open-source molecular docking program that integrates deep learning to enhance the accuracy of protein-ligand binding predictions. GNINA is a fork of smina and AutoDock Vina, and its primary innovation is the use of convolutional neural networks (CNNs) to improve pose prediction and scoring. This approach leads to more reliable virtual screening and drug discovery workflows by outperforming traditional docking methods on redocking and cross-docking tasks.
How GNINA Works
GNINA combines the strengths of traditional docking with modern deep learning methods to create a powerful and accurate tool:
Hybrid Scoring: In its default mode, GNINA uses Monte Carlo sampling to generate ligand poses, which are initially scored with the traditional Vina empirical scoring function. However, the final poses are re-ranked using an ensemble of CNN models that have been trained on thousands of protein-ligand complexes.
Enhanced Performance: This CNN-based scoring significantly improves docking performance. In one benchmark, GNINA increased the percentage of targets where the top pose was within 2Å of the known binding pose from 58% to 73% in redocking tasks and from 27% to 37% in cross-docking tasks, compared to AutoDock Vina.
Versatile and Flexible: GNINA supports a variety of docking scenarios, including docking with flexible side-chain residues and whole-protein docking. It also allows for customizable scoring functions, giving researchers the flexibility to use a combination of empirical and machine learning-based options.
What is Tamarind Bio?
Tamarind Bio is a pioneering no-code bioinformatics platform built to democratize access to powerful computational tools for life scientists and researchers. Recognizing that many cutting-edge machine learning models are often difficult to deploy and use, Tamarind provides an intuitive, web-based environment that completely abstracts away the complexities of high-performance computing, software dependencies, and command-line interfaces.
The platform is designed provide easy access to biologists, chemists, and other researchers who may not have a background in programming or cloud infrastructure but want to run experimental models with their data. Key features include a user-friendly graphical interface for setting up and launching experiments, a robust API for integration into existing research pipelines, and an automated system for managing and scaling computational resources. By handling the technical heavy lifting, Tamarind empowers researchers to concentrate on their scientific questions and accelerate the pace of discovery.
Accelerating Discovery with GNINA on Tamarind Bio
Using GNINA on a platform like Tamarind could dramatically accelerate drug discovery by providing a highly accurate, integrated, and user-friendly workflow.
Improved Virtual Screening: GNINA outperforms conventional empirical scoring functions in virtual screening benchmarks, with a median 1% early enrichment factor that is more than twice that of Vina. This suggests that using GNINA on Tamarind.bio would allow researchers to more effectively identify promising drug candidates from large compound libraries.
High-Quality Docking: The platform could leverage GNINA's improved pose prediction accuracy to provide more reliable docking results, enabling researchers to make more informed decisions during hit-to-lead optimization.
Integrated Workflow: A platform like Tamarind would integrate GNINA into a seamless workflow, automating the process of preparing protein and ligand structures, running the docking simulation, and providing a ranked list of poses based on the CNN-based scoring function.
How to Use GNINA on Tamarind Bio
To leverage GNINA's power, a researcher could follow this streamlined workflow on Tamarind:
Access the Platform: Begin by logging in to the tamarind.bio website.
Select GNINA: From the list of available computational models, choose the GNINA tool.
Input Structures and Binding Site: Provide the PDB file of the receptor protein and the ligand in an SDF or PDB file format. Define the binding site by drawing a box around a known ligand pose or by providing an entire protein structure for blind docking.
Specify Docking Parameters: Choose from various docking options, including whether to use flexible side-chains or different CNN scoring methods (e.g., rescore, refinement, all).
Run Docking Simulation: The platform would run GNINA, which performs Monte Carlo sampling to generate poses and then uses its CNN-based scoring function to re-rank them.
Analyze and Select: The final output would include a list of ranked poses with corresponding scores. You can use this information to select the most favorable binding conformations for further analysis.