STELLAR
Advanced docking protocol for large flexible peptides — without AI.
About
STELLAR is a docking protocol developed at BIO-HPC for the docking of large peptides and other large, flexible ligands beyond the typical small-molecule range. It does not rely on AI methods, using instead a fragment-and-assemble strategy with HPC-distributed search.
Originally introduced in a cancer-context application, it is suited for problems where peptide ligands or large flexible biopolymers (e.g. glycans, polysaccharides) need accurate pose prediction.
Lead authors: Alejandro Rodríguez Martínez
Key features
Large-peptide docking
Beyond the small-molecule limits of conventional docking programs.
Fragment-based assembly
Splits flexible ligands into manageable fragments and reassembles consensus poses.
No AI dependency
Deterministic, physics- and geometry-based, useful where AI training data is scarce.
Access
HPC suite
Read the preprint (chemRxiv 2025)
Preprint released · public release in preparation
https://doi.org/10.26434/chemrxiv-2025-nmn5s →Cite this tool
STELLAR: developing and optimising a novel advanced docking protocol for processing large peptides without AI assistance — a cancer-context application