There is a newer version of the record available.

Published July 8, 2026 | Version v0.3.1

Lacuna: Cryptic Binding Pocket Discovery via Conformational Ensemble Analysis

Authors/Creators

Description

Lacuna discovers cryptic binding pockets in proteins by generating a conformational ensemble, detecting pockets per conformer, clustering across the ensemble, and ranking by a continuous crypticity score. Evaluated on 22 cryptic-pocket apo/holo pairs (7/22, 32% under a size-robust Jaccard-or-centroid criterion; 59% under the legacy size-gameable recall metric). Uses an Anisotropic Network Model backend on CPU (~1-8 s/protein).

Notes

If you use Lacuna in your research, please cite it as below.

Files

mooreneural/lacuna-v0.3.1.zip

Files (8.6 MB)

Name Size Download all
md5:3038a3b4cad4bac7fd0a8933cad3f72b
8.6 MB Preview Download

Additional details

Related works