A framework for conformational ensemble generation of disordered regions integrating structural database sampling and AlphaFold predictions
Year: 2025
Type: article
Abstract: "Intrinsically disordered proteins (IDPs) lack stable structures and exist as dynamic conformational ensembles. Even groundbreaking methods of protein structure prediction, like AlphaFold, have limitations in predicting IDP conformational properties, so these remain poorly understood. Capturing their dynamic nature requires instead generating structural ensembles, a computationally intensive task that often requires expert knowledge. Existing approaches for ensemble generation like structural database sampling (SDS), molecular dynamics (MD), and artificial intelligence/machine learning (AI/ML) methods have limitations. MD and AI/ML methods are computationally expensive and require significant expertise, while SDS methods either prioritize accessibility, lacking versatility or customization, or offer flexibility with complex setups, becoming inaccessible for inexperienced researchers. Additionally, current methods struggle to model intrinsically disordered regions in multidomain or multichain proteins, don’t account for non-random secondary structure propensities in IDPs and rarely exploit information stemming from AlphaFold’s output.(...)"
Author: Pires Fernandes, Nuno Miguel
Language: English
Subfield: Molecular Biology
Domain: Life Sciences
Open Access status: green
OpenAlex ID: W7163291097