✳flâneur — a map of the web's best reading

Strengths and limitations of AlphaFold 2 | AlphaFold

ebi.ac.uk · 1,390 words · saved by 1 readers

All materials are free cultural works licensed under a Creative Commons Attribution 4.0 International (CC BY 4.0) license, except where further licensing details are provided. Share this page with: What a neural network can and can’t do largely depends on the data used for its training. For training AlphaFold2, only the protein parts of the PDB structures were used: other parts like small molecules and nucleic acids were not included in the training. This means there are some aspects of the structures that AlphaFold cannot predict, or cannot guarantee accuracy. AlphaFold2 was originally trained on single protein chains, so it excels at predicting their structures. Later, an extension of AlphaFold2 was trained specifically to predict protein-protein complexes: this version is now known as AlphaFold-Multimer (Evans et al., 2022). It can predict the structures of protein complexes that are made of several copies of the same chain (homo-multimers such as dimers and hexamers) as well as tho

Strengths and limitations of AlphaFold 2 | AlphaFold --> --> AlphaFold A practical guide Course progress: 0% Open Tree Course overview Search within this course An introductory guide to AlphaFold’s strengths and limitations Open Tree What are proteins and how do we know their structures? What is the protein folding problem? What is AlphaFold? Strengths and limitations of AlphaFold 2 Test your knowledge Validation and impact Open Tree How have AlphaFold2’s predictions of protein structure been validated? How accurate are AlphaFold 2 structure predictions? How is AlphaFold 2 used by scient

Explore this link on the map →

related reading