- Course overview
- Search within this course
- An introductory guide to AlphaFold’s strengths and limitations
- Validation and impact
- Inputs and outputs
- Accessing and predicting protein structures with AlphaFold 2
- Choosing how to access AlphaFold2
- Accessing predicted protein structures in the AlphaFold Database
- Predicting protein structures with ColabFold and AlphaFold2 Colab
- Predicting protein structures using the AlphaFold2 open-source code
- Other ways to access predicted protein structures
- How to cite AlphaFold
- Advanced modelling and applications of predicted protein structures
- Classifying the effects of missense variants using AlphaMissense
- AlphaFold 3 and AlphaFold Server
- Summary
- Course slides
- Your feedback
- Glossary of terms
- Acknowledgements
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.
Summary
AlphaFold 3 represents a significant leap forward in our ability to understand the molecular world. By predicting the structures of complexes encompassing a vast array of biomolecules and their interactions, it opens up new avenues for research and discovery across multiple disciplines.
Key takeaways:
We’ve explored how AlphaFold 3 extends beyond protein structure prediction to model intricate biological assemblies involving DNA, RNA, ligands, ions, and diverse chemical modifications.
We’ve seen how the innovative architecture of AlphaFold 3 leads to more accurate structure predictions, in many cases even surpassing specialised methods.
We’ve delved into the practical use of AlphaFold 3 via AlphaFold Server, including both generating structure predictions and critically interpreting them.
The broader impact and ongoing evolution of AlphaFold
AlphaFold has revolutionised the field of protein structure prediction. It predicts the 3D structures of proteins with remarkable accuracy, based solely on the amino acid sequence. (See “Introductory guide“). AlphaFold’s capabilities have been validated by multiple lines of evidence from structural biology experiments, including studies of X-ray crystallography, cryogenic electron microscopy, and cross-linking mass spectrometry. (See “Validation and Impact“).
Prediction of missense variant effects
Missense variants are the most common type of genetic variant. They involve a single change in the DNA sequence that results in a substitution of one amino acid for another in a protein. While some missense variants are harmless, others can lead to genetic disorders.
To predict the effects of missense variants, Google DeepMind has built a new tool called AlphaMissense. Based on AlphaFold 2, AlphaMissense is a separate system that can predict whether a missense genetic variant is likely to be pathogenic (disease-causing) or benign (limited effect).
To achieve this, AlphaMissense analyses a massive dataset of variants. This includes information about each variant’s frequency in the human population and their location in the protein sequence. Crucially, it uses structural context derived from AlphaFold to predict the effects of the missense variants on protein structure and function.
AlphaMissense has already provided high-confidence predictions for most human missense variants (Cheng et al., 2023).
The future of AlphaFold and its impact
The journey of AlphaFold 3 has just started. As researchers continue to explore its capabilities and limitations, we anticipate groundbreaking discoveries in structural biology, biochemistry, plant sciences, and beyond. Furthermore, AlphaFold Server is currently under development, and we expect more features to emerge soon.
AlphaFold can provide testable hypotheses that guide experiments. It has been applied in a wide range of areas, including guiding mutational analyses, analysis of protein action and potential interactions, protein engineering, and finding distantly homologous structures. Fields as diverse as drug discovery, computational biology, and biotechnology are likely to see significant advances thanks to AlphaFold.
AlphaFold is not just a technological breakthrough, but a catalyst for scientific innovation. The open-source release of AlphaFold has fostered a wave of advances within the scientific community, leading to the development of new applications and tools. The Google DeepMind team is thrilled to witness these creative and groundbreaking applications of AlphaFold, and eagerly anticipates future contributions from the community.
We stand at the precipice of a transformative era in biological research. AlphaFold is propelling us towards a future where protein structure prediction is no longer a challenge, but instead a ubiquitous tool for unlocking new discoveries and innovations.
Empowering researchers
Through this short course, you’ve gained the knowledge and skills to harness the power of AlphaFold 2 and 3 in your own research. Whether you’re investigating protein function or exploring the intricacies of molecular interactions, AlphaFold 3 has the potential to accelerate your work and unlock new insights.
Additional resources for further learning
To find out about more opportunities to join live courses and on-demand training at EMBL-EBI, sign up to their newsletter(s).
To stay up-to-date on the latest developments from Google DeepMind, including AlphaFold, follow the Google DeepMind blog or visit the Google DeepMind website.