Peptide–protein docking

What is peptide–protein docking?

Peptide–protein docking is a computational method for placing a peptide on a protein receptor and generating plausible bound conformations. The search is harder than typical small-molecule docking because peptide backbones and side chains can adopt many conformations.

A peptide has many backbone and side-chain degrees of freedom and may bind in a conformation that is rare in solution. Docking must therefore sample peptide shape, receptor location, and orientation together, making the problem harder than placing a comparatively rigid small molecule.

Known sites and motifs support a focused, restrained search over selected peptide conformers. Unknown sites require broader surface search and more conformational sampling. Post-translational modifications, terminal charges, cyclization, and noncanonical residues must be represented explicitly because each can change peptide geometry and receptor contacts.

When to use peptide–protein docking

• Suitable peptide–protein docking question. Modeling peptide binders, linear motifs, and transient signaling interfaces
• Required structures and evidence are available. A prepared receptor and peptide structure or a defensible peptide conformational model

Benefits of peptide–protein docking

• Practical output. Handles a biologically important interaction class
• Comparative evidence. Can incorporate motif or site restraints
• Connected analysis. Returns explicit peptide contact hypotheses

Primary limitations

• Method dependence. Peptide flexibility expands the search space
• Input sensitivity. Unbound peptide conformations may be misleading
• Validation boundary. Scores do not establish cellular activity

How peptide–protein docking works

Peptide docking methods trade search breadth against the amount of prior information supplied.

• Local restrained docking. Known receptor residues or sequence motifs focus the calculation on a defined site. This is efficient but can only recover solutions compatible with the supplied restraints.
• Global protein–peptide docking. LightDock and related methods explore the receptor surface when the site is uncertain. Multiple starting peptide conformers improve coverage but increase cost.
• Flexible or learned complex modeling. Flexible refinement and learned protein-complex methods can adjust peptide geometry, although confidence may be lower for unusual modifications or conformations.

Applications of peptide–protein docking

Protein–peptide docking helps interpret compact recognition motifs and design experiments around transient interfaces.

• Motif recognition. Model how a linear motif, degron, cleavage-region peptide, or signaling segment may occupy a receptor groove.
• Peptide binder design. Compare residue substitutions or constrained peptide concepts before synthesis and experimental testing.
• Interface validation. Propose receptor and peptide contacts for alanine scanning, competition experiments, or structural follow-up.

How to do peptide–protein docking online

The online workflow accepts explicit receptor and peptide structures, making peptide conformational assumptions visible before docking.

1. Prepare the receptor. Upload the biologically relevant PDB structure, select the correct chain and state, and inspect the proposed binding region for missing residues or cofactors.
2. Build peptide conformations. Create one or more starting peptide structures with the correct sequence, stereochemistry, terminal state, cyclization, and modifications.
3. Define available restraints. Record motif positions, receptor residues, cross-links, or competition data. Use uncertain evidence as a comparison condition rather than a hard fact.
4. Run HADDOCK3 and LightDock. Compare information-driven and global swarm-based docking from the same receptor and peptide inputs, retaining all clusters and method-native scores.
5. Cluster and inspect complexes. Review peptide backbone geometry, recurring contacts, restraint satisfaction, clashes, and dependence on the starting conformer before export.

How to interpret peptide–protein docking results

A peptide pose is more credible when it recurs across starting conformers or methods, satisfies independent residue evidence, and adopts physically plausible backbone and side-chain geometry. A single favorable score can reflect an overfit peptide conformation.

Validate predicted contacts with peptide substitution series, receptor mutagenesis, competition or binding assays, and structural data. Docking alone does not establish cellular uptake, proteolytic stability, selectivity, or functional activity.

How the peptide–protein docking workflow works

Dock the same peptide and receptor with HADDOCK3 and LightDock, then inspect method-specific complexes.

1. Prepare the receptor. Upload the biologically relevant PDB structure, select the correct chain and state, and inspect the proposed binding region for missing residues or cofactors.
2. Build peptide conformers. Create one or more starting peptide structures with the correct sequence, stereochemistry, terminal state, cyclization, and modifications.
3. Add binding-site evidence. Record motif positions, receptor residues, cross-links, or competition data. Use uncertain evidence as a comparison condition rather than a hard fact.
4. Dock and refine. Compare information-driven and global swarm-based docking from the same receptor and peptide inputs, retaining all clusters and method-native scores.
5. Cluster and inspect. Review peptide backbone geometry, recurring contacts, restraint satisfaction, clashes, and dependence on the starting conformer before export.

Inputs and outputs

Check formats before running, then inspect and download the result from every workflow step.

Inputs

• Structural inputs. PDB SDF SMILES A receptor PDB and one or more peptide PDB conformations, optionally with binding-site restraints.
• Method context. Binding-site evidence, restraints, receptor-state provenance, known ligands, or reference complexes when available.

Outputs

• Docked structures. PDB PDBQT SDF Ranked protein–peptide complexes, clusters, residue contacts, and method-native scores.
• Review evidence. Method-native rankings, confidence, logs, interaction context, failures, and files for reproducible follow-up.