🧬 How to use DiffDock-L Online

Use DiffDock-L online for blind protein-ligand docking and confidence-ranked pose generation.

DiffDock-L is a diffusion-based protein-ligand docking model built to improve blind docking on unseen pockets and protein classes. The accompanying paper focuses on generalization: it introduces the DockGen benchmark, shows that earlier machine-learning docking models generalize poorly across the proteome, and demonstrates that larger models, larger datasets, synthetic data, and improved confidence training materially raise out-of-distribution performance.

On Neurosnap, researchers submit an Input Receptor and Input Ligand and choose Number Samples to control how many poses are explored. The workflow is useful for early pose generation, library triage, and rapid comparison of alternative binding modes when a structural hypothesis is needed before more expensive physics-based follow-up.

How DiffDock-L Works

DiffDock-L extends the original DiffDock formulation with a stronger focus on confidence-guided generalization. The paper highlights Confidence Bootstrapping, a training strategy that couples diffusion generation with confidence prediction across the multi-resolution denoising process, helping the model recognize which poses are trustworthy on novel protein families.

On Neurosnap, the main scientific question is not whether a single top pose exists, but whether the sampled pose family converges. The learned DiffDock-L confidence scores are presented together with complementary energy-style checks from the platform's secondary scoring stage, giving researchers a more defensible basis for deciding which complexes deserve molecular dynamics, medicinal chemistry review, or experimental validation.

What is Neurosnap?

Neurosnap is the leading platform for bioinformatics and computational science focused on expanding access to powerful modeling and simulation tools. Because many state-of-the-art machine learning systems remain complex to install, configure, and scale, Neurosnap offers a clean, browser-based workspace that removes the burden of infrastructure management, dependency conflicts, and command-line tooling.

Built for biologists, chemists, and cross-disciplinary scientists, the platform enables advanced computational workflows without requiring expertise in software engineering or cloud architecture. Researchers can launch analyses through an intuitive interface, connect programmatically through a comprehensive API, and rely on automated resource management to scale workloads efficiently. By taking care of the underlying compute and operational complexity, Neurosnap allows teams to devote their energy to scientific progress and faster iteration. Security and data protection remain foundational principles, with clear safeguards outlined in our Terms of Use and Privacy Policy to ensure your work stays protected.

Advancing Discovery with DiffDock-L on Neurosnap

Using DiffDock-L on Neurosnap could drastically accelerate blind protein-ligand docking with learned confidence ranking for unseen binding pockets.

Blind docking workflow: DiffDock-L starts from a receptor structure and ligand alone, making it practical when the binding site is uncertain or only weakly characterized.
Generalization focus: DockGen and confidence bootstrapping were introduced specifically to improve performance on unseen targets rather than only on familiar benchmark pockets.
Search breadth control: Number Samples lets researchers decide how exhaustively to explore alternative poses for a given receptor-ligand pair.
Pose-family interpretation: Learned confidence plus secondary energy checks make it easier to distinguish stable pose families from isolated, less trustworthy predictions.

How to Use DiffDock-L on Neurosnap

To harness the capabilities of DiffDock-L, researchers can follow this streamlined workflow within Neurosnap:

Access Neurosnap: Start by logging in to the Neurosnap website.
Select Tool: From the list of available tools, choose DiffDock-L.
Provide Inputs: Provide all the inputs specified within the submission panel and optionally configure the tool as desired.
Run Tool: Submit the DiffDock-L job and Neurosnap will execute it in the cloud, automatically notifying you as soon as your results are ready.
Review Output: Explore your results through rich visualizations, including figures, plots, and interactive views designed to help you analyze findings with clarity and confidence.

Similar Services

Explore related tools that support similar research workflows:

AutoDock Vina (smina)

DynamicBind

GNINA

SPRINT

Proudly supporting 50,000+ scientists worldwide, including 7,000+ leading biotech and global biopharma organizations.