New AI solution that can determine the shape of proteins could 'revolutionize' medical research

DeepMind’s artificial intelligence (AI) solution, AlphaFold, has succeeded in determining the shape of proteins with a level of accuracy similar to that of laboratory experiments. The breakthrough will be instrumental in furthering our knowledge of human health and disease, helping us to develop new diagnostics and therapeutics.

Artificial Intelligence

Image Credit: Blue Planet Studio/

Developing AI to read proteins

Researchers at the 14th Community Wide Experiment on the Critical Assessment of Techniques for Protein Structure Prediction (CASP14) have announced that they have successfully developed an AI platform capable of reading the shape of proteins, a task vital to understanding the nature of human disease which is intrinsically linked with the way proteins function.

Our DNA codes for amino acid sequences that form the tens to possibly hundreds of thousands of proteins that exist in the human body. Research has shown that almost every kind of human disease is linked with the function of these proteins and that each protein’s function is related to its shape. For more than half a century researchers have endeavored to establish a rapid and accurate method of reading a protein’s shape to know more about the human body and the diseases that inflict it.

Currently, determining the shape of each given protein requires a significant amount of resources, including expensive equipment and the work of skilled scientists over months or even years. With hundreds of thousands of proteins potentially of interest to scientists studying disease, the scientific community has been in great need of an enhanced system.

DeepMind, an AI lab based in London, took part in the CASP challenge where research teams from around the world were given sequences of amino acids for a set of roughly 100 proteins. The researchers were then tasked with developing a computer-based system to determine the shape of the proteins to the same level of accuracy as the time-consuming lab experiments.

The results of the study, which were determined by independent scientists, revealed that DeepMind's AlphaFold program was effective at establishing the shape of roughly two-thirds of the proteins with a similar level of accuracy to that of the lab experiments. In addition, the program’s accuracy at decoding the rest of the proteins was also high, but just not to the same level as the lab tests.

The organizers of CASP stress how DeepMinds achievement builds on those made in previous CASP rounds, helping to significantly advance this field of science. With the AI that DeepMind has built, scientists will be much better equipped at understanding the shape of structures with speed, something that will be vital to future medical research.

Revolutionizing medical research

Dr. Kryshtafovych of UC Davis, USA, and CASP organizer states that "being able to investigate the shape of proteins quickly and accurately has the potential to revolutionize life sciences. Now that the problem has been largely solved for single proteins, the way is open for development of new methods for determining the shape of protein complexes - collections of proteins that work together to form much of the machinery of life, and for other applications.”

Up until this point, the mystery of how proteins fold to create highly complex and unique three-dimensional structures had remained largely unsolved. Now, with the work done by DeepMind and CASP, scientists have an opportunity to gain further insight into how these unique shapes help the protein function on a molecular level.

The future of our understanding of human health and disease will benefit greatly from the advancements made by DeepMind’s new AI program. It will also inspire continued research in this field, which will undoubtedly lead to improvements in diagnostic, preventative, and therapeutic approaches to disease.

Sarah Moore

Written by

Sarah Moore

After studying Psychology and then Neuroscience, Sarah quickly found her enjoyment for researching and writing research papers; turning to a passion to connect ideas with people through writing.


Please use one of the following formats to cite this article in your essay, paper or report:

  • APA

    Moore, Sarah. (2020, November 30). New AI solution that can determine the shape of proteins could 'revolutionize' medical research. News-Medical. Retrieved on July 24, 2024 from

  • MLA

    Moore, Sarah. "New AI solution that can determine the shape of proteins could 'revolutionize' medical research". News-Medical. 24 July 2024. <>.

  • Chicago

    Moore, Sarah. "New AI solution that can determine the shape of proteins could 'revolutionize' medical research". News-Medical. (accessed July 24, 2024).

  • Harvard

    Moore, Sarah. 2020. New AI solution that can determine the shape of proteins could 'revolutionize' medical research. News-Medical, viewed 24 July 2024,


  1. Warren Hall Warren Hall United States says:

    It would have been easy to go far overboard in estimating the future yields of this AI application. As it is, the article is well written, balanced, fair, and grammatically correct. A cut above many science and technology writers!

The opinions expressed here are the views of the writer and do not necessarily reflect the views and opinions of News Medical.
Post a new comment

While we only use edited and approved content for Azthena answers, it may on occasions provide incorrect responses. Please confirm any data provided with the related suppliers or authors. We do not provide medical advice, if you search for medical information you must always consult a medical professional before acting on any information provided.

Your questions, but not your email details will be shared with OpenAI and retained for 30 days in accordance with their privacy principles.

Please do not ask questions that use sensitive or confidential information.

Read the full Terms & Conditions.

You might also like...
Breakthrough in aging research: Blocking IL-11 extends lifespan and improves health in mice