Related Experiment Video
Updated: Jun 10, 2026

A Protocol for Computer-Based Protein Structure and Function Prediction
Published on: November 3, 2011
Regularly updated benchmark sets for statistically correct evaluations of AlphaFold applications
Laszlo Dobson1,2, Gábor E Tusnády1,2, Peter Tompa1,3,4
1Institute of Molecular Life Sciences Research, Centre for Natural Sciences, Magyar Tudósok Körútja, Budapest, Hungary.
Abstract:
AlphaFold2 changed structural biology by providing high-quality structure predictions for all possible proteins. Since its inception, a plethora of applications were built on AlphaFold2, expediting discoveries in virtually all areas related to protein science. In many cases, however, optimism seems to have made scientists forget about data leakage, a serious issue that needs to be addressed when evaluating machine learning methods. Here we provide a rigorous benchmark set that can be used in a broad range of applications built around AlphaFold2/3.
Related Concept Videos
Protein Folding Quality Check in the RER
Expected Frequencies in Goodness-of-Fit Tests
Bonferroni Test
The means of different samples are first paired in all possible combinations.
The null hypothesis of the...

