Work

Realism and resolution: How AI-predicted protein structures differ from experimental structures

Year: 2026

Type: preprint

Abstract: Protein structure predictors can recover protein folds with remarkable accuracy, and their models are increasingly used as if they were experimental structures. Yet even for a confidently predicted fold, it remains difficult to judge how faithfully the model reproduces protein geometry. We introduce WhatRes, which assesses a structure from its coordinates on two complementary axes. Apparent resolution estimates the experimental resolution its local geometry most closely resembles. Realism measures how closely that geometry matches experimental structures at the same apparent resolution. Predicted models can score well on apparent resolution while showing systematically less variation in bond lengths and angles than is observed in high-resolution crystal structures. WhatRes reveals this over-idealisation even in high-confidence models.

Source: bioRxiv (Cold Spring Harbor Laboratory)

DOI: 10.64898/2026.11.06.764824

Authors: Rob van der Kant, Javier Delgado Blanco, Luis Serrano, Joost Schymkowitz, Frederic Rousseau

Institutions: VIB-KU Leuven Center for Brain & Disease Research, Barcelona Institute of Science and Technology, Centre for Genomic Regulation, KU Leuven

Language: English

Topic: Protein Structure and Dynamics

Subfield: Molecular Biology

Field: Biochemistry, Genetics and Molecular Biology

Domain: Life Sciences

Open Access status: green

OpenAlex ID: W7225645430

Cites:

Cited by:

Related to: