The LiPP Benchmark Set for Modeling Lipid–Protein Complexes: Comparison of Co-Folding and Docking Methods
Li-Yen Yang, Shreyas Gupta, Lauren N. Mullininx, Andrew C. McShan · 2026
Read The LiPP Benchmark Set for Modeling Lipid–Protein Complexes: Comparison of Co-Folding and Docking Methods by Li-Yen Yang, Shreyas Gupta, Lauren N. Mullininx, Andrew C. McShan (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/PMC13292216.txt.
Description
Li-Yen Yang and co-authors introduce a 2026 benchmark called LiPP for testing how well software predicts lipid-protein binding poses. Predicting where lipids dock on proteins trails other complex types, partly for lack of shared ground-truth data. The team curated 311 nonredundant structures spanning many protein folds and lipid classes, then ran five modeling tools against them.
Three AI co-folding tools, one AI docking tool, and one physics-based docker were compared. AlphaFold 3 scored best, with 76.1 percent of poses under the accuracy cutoff, though that fell to 47.2 percent on structures outside its training data. Physics-based docking and some AI tools kept poses physically plausible, while others broke basic constraints. The authors offer LiPP as a standard test bed and call for better handling of lipid size and flexibility.
Catalog record
| CREATOR | Li-Yen Yang, Shreyas Gupta, Lauren N. Mullininx, Andrew C. McShan |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 80465 |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |