How to use zPocket Online

Use zPocket online to predict ligand-binding pockets from a single protein sequence.

zPocket predicts ligand-binding residues from frozen pair representations of ESMFold2. Provide one standard amino-acid sequence and choose a generated MSA, upload a custom A3M, or use single-sequence mode. It folds the monomer, scores each residue, and clusters the scores into ranked pocket centres. It does not score an uploaded experimental structure.

The head was trained on single-chain monomer folds and cannot place pockets that exist only at chain interfaces. Inspect structural confidence before interpreting sites in uncertain regions.

How zPocket Works

The service uses the released ESMFold2 pairformer2 head and its published folding recipe: 10 refinement loops, language-model dropout of 0.25, 100 sampling steps and three diffusion samples. The first conformer is returned and used to locate pocket centres. The default decoder uses score-weighted mean shift with a probability threshold of 0.3, a bandwidth of 14 Å and at least three contributing residues.

residue_scores.csv contains 1-based residue positions, amino acids, pocket probabilities and pLDDT on a 0–100 scale. pockets.csv lists ranks, summed probability scores, mean probabilities, residue positions and centres in the returned fold's coordinate frame. summary.csv reports sequence length, pocket count and confidence summaries. fold.cif retains pLDDT; pocket_probabilities.cif stores probabilities on a 0–1 scale in the B-factor column. These are two views of the same predicted structure.

Probabilities rank residues and are not a universal binding cutoff or a druggability measurement. Pocket scores are summed residue probabilities, so larger groups can score higher. When no residue passes the threshold, the decoder uses the top ten residues and marks the output as fallback. Zero retained pockets is also a valid result. Predictions are less reliable in low-pLDDT regions, and single-sequence runs are weaker than MSA-conditioned predictions. Centres refer only to the returned fold and need alignment before use with other structures.

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 zPocket on Neurosnap

Using zPocket on Neurosnap could drastically accelerate sequence-based ligand-binding pocket prioritization with residue scores and structural context.

Sequence input: Find candidate sites without supplying an experimental structure.
Residue-level evidence: Inspect pocket probabilities together with fold confidence.
Ranked spatial sites: Use pocket centres and contributing residues to guide structural review and follow-up studies.

How to Use zPocket on Neurosnap

To harness the capabilities of zPocket, 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 zPocket.
Provide Inputs: Provide all the inputs specified within the submission panel and optionally configure the tool as desired.
Run Tool: Submit the zPocket 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.

Citations

Please cite the original work when using zPocket in publications or research outputs.

Portnoi, T., Keys, A. M., DiFrancesco, B. R., & Gruver, N. (2026). zPocket: Better Pocket Finding with Frozen Co-folding Features. arXiv:2610.17696.

Neurosnap Inc. (2022). Neurosnap: An online platform for computational biology and chemistry. Available at: https://neurosnap.ai/

Similar Services

Explore related tools that support similar research workflows:

AF2Bind

Lacuna Cryptic Pocket Discovery

SPRINT

AllMetal3D

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