About

Predicting the 3D structure of proteins is one of the fundamental grand challenges in biology. By solving this challenge, we can dramatically deepen our understanding of human health, disease, and our environment, especially within areas like drug design and sustainability.


AlphaFold, the state-of-the-art AI system developed by Google DeepMind, is able to computationally predict protein structures with unprecedented accuracy and speed. Working in partnership with EMBL’s European Bioinformatics Institute (EMBL-EBI), we’ve released over 200 million protein structure predictions by AlphaFold that are freely and openly available to the global scientific community. Included are nearly all catalogued proteins known to science – with the potential to increase humanity’s understanding of biology by orders of magnitude.

Explore stories of AlphaFold’s impact

Paper & methodologies

The paper describing the AlphaFold method is:
Jumper, J et al. Highly accurate protein structure prediction with AlphaFold (Nature 2021).


For a discussion of AlphaFold’s output when applied to a whole proteome, see:
Tunyasuvunakool, K et al. Highly accurate protein structure prediction for the human proteome (Nature 2021).

What are proteins?

Inside every cell in your body, billions of tiny molecular machines are hard at work. They’re what allow your eyes to detect light, your neurons to fire, and the ‘instructions’ in your DNA to be read, which make you the unique person you are.

These exquisite, intricate machines are proteins. They underpin not just the biological processes in your body but every biological process in every living thing. They’re the building blocks of life.

What are protein structures?

Currently, there are around 230 million proteins in UniProt, with millions more catalogued every year. Each one has a unique 3D shape that determines how it works and what it does. AlphaFold DB now offers a prediction of what each of the over 200 millions known proteins in UniProt might look like.

If you could unravel a protein you would see that it’s like a string of beads made of a sequence of different chemicals known as amino acids. These sequences are assembled according to the genetic instructions of an organism's DNA.

Attraction and repulsion between the 20 different types of amino acids cause the string to fold in a feat of ‘spontaneous origami’, forming the intricate curls, loops, and pleats of a protein’s 3D structure.

For decades, scientists have been trying to find a method to reliably predict a protein’s structure just from its sequence of amino acids.

 This grand scientific challenge is known as the protein folding problem.

Why does being able to predict a protein’s 3D structure matter?

In the last 60 years, the scientific community has been using advanced experimental methods to determine the structures of over 180,000 proteins in atomic detail. This work has already improved our understanding of many fundamental processes in health and disease:

  • Determining the structure of haemoglobin, the protein in red blood cells responsible for carrying oxygen around the body, helped researchers understand how a single mutation can cause sickle-cell anaemia, helping develop treatments for the condition.
  • Determining the structure of the SARS-CoV-2 viral proteins enabled scientists to understand how it operates and to identify treatments and develop new vaccines.
  • Determining the structure of the photosynthetic reaction centre improved researchers’ understanding of how photosynthesis works in plants.


But figuring out the exact structure of a protein remains an expensive and often time-consuming process, meaning we only know the exact 3D structure of a tiny fraction of the proteins known to science.

How can AI and AlphaFold help?

Finding a way to close this rapidly expanding gap and predict the structure of millions of unknown proteins could not only help us tackle disease and more quickly find new medicines but ultimately better understand life itself. This is why biologists are turning to AI methods: the ability to predict a protein’s shape computationally from its genetic code alone – as a complementary alternative to determining it through costly and time-consuming experimentation – could help dramatically accelerate research.

AlphaFold is an AI system developed by Google DeepMind that makes state-of-the-art accurate predictions of a protein’s structure from its amino-acid sequence. In 2020, AlphaFold was recognised as a solution to the protein folding problem by the organisers of the CASP14 benchmark, a biennial challenge for research groups to test the accuracy of their predictions against real experimental data.

AlphaFold’s unprecedented accuracy and speed has enabled the creation of an extensive database of structure predictions, and opens up the potential for scientists to use computational structure prediction as a core tool in their research.

Visit our online training course

How can we make the most of the data?

A ‘mountain’ of new scientific data is not necessarily useful. To extract knowledge from scientific data you have to be able to search it in a variety of ways, to compare it and to analyse it.

EMBL-EBI is a global leader in the storage, analysis and dissemination of large biological research datasets. The institute manages over 40 biological data resources, used by the global scientific community. On any given day, millions of scientists use these data resources to advance their work.

Building on its big-data expertise, EMBL-EBI adds value to the AlphaFold dataset by curating and organising it, and linking it to other biological data resources such as the PDB and UniProt. This way, scientists can explore the structure of their protein of interest in a wider biological context. Curating the data in this way increases its value and usefulness - similar to the difference between a tonne of bricks and a house.

The AlphaFold Protein Structure Database will continue to be improved and expanded in the future, adding more protein structures and functionalities. If you would like to learn more, you can read our blog post about the AlphaFold DB.

About Google DeepMind

Google DeepMind is a scientific discovery company, committed to ‘building AI responsibly to benefit humanity.’ Achieving this goal requires a diverse and interdisciplinary team working closely together – from scientists and designers, to engineers and ethicists – to pioneer the development of advanced artificial intelligence.

The company’s breakthroughs include AlphaGo, AlphaFold, more than 1,000 publications. By solving some of the hardest scientific and engineering challenges of our time, GDM is working to create breakthrough technologies that could advance science, transform work, serve diverse communities — and improve billions of people’s lives.

About EMBL-EBI

EMBL-EBI is a not-for-profit international institute that helps scientists realise the potential of big data. The institute collaborates with scientists and engineers all over the world, and provides the infrastructure needed to share data openly and fairly in the life sciences. It also performs computational research and delivers bioinformatics training for the global scientific community. EMBL-EBI is part of the European Molecular Biology Laboratory (EMBL).