Benchmarking sequence-based and AlphaFold-based methods for pMHC-II binding core prediction: distinct strengths and consensus approaches
Soobon Ko, Honglan Li, Hongeun Kim, Woong-Hee Shin, Junsu Ko, Yoonjoo Choi · 2026
Read Benchmarking sequence-based and AlphaFold-based methods for pMHC-II binding core prediction: distinct strengths and consensus approaches by Soobon Ko, Honglan Li, Hongeun Kim, Woong-Hee Shin, Junsu Ko, Yoonjoo Choi (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/PMC13410576.txt.
Description
Soobon Ko and colleagues benchmark in 2026 computer tools that predict where a peptide sits in an immune display molecule. Those molecules show peptide pieces to T cells, and the grip point, called the binding core, shapes the immune answer. Getting the core right matters for vaccine and therapy design.
Four tools were tested on 72 known complexes plus shuffled and known non-binding peptides. Two structure-based tools built on AlphaFold caught most true binders yet often called non-binders positive. A sequence-based tool, NetMHCIIpan, proved best at spotting non-binders while missing more weak positives. Blending tools into agreed calls balanced those strengths, and the authors map when each blend suits best.
Catalog record
| CREATOR | Soobon Ko, Honglan Li, Hongeun Kim, Woong-Hee Shin, Junsu Ko, Yoonjoo Choi |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 41545 |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |