You may have heard about "DeepMind" in the past, and if you haven’t, now you will.
To this day, DeepMind has acquired a number of achievements since it was founded, but it is most notable for AlphaGo, an AI program that beat some of the best professional Go players in history including Ke Jie.
The Big News
Today, DeepMind has outdone itself again by solving one of the biggest challenges in biology:
DeepMind’s AlphaFold 2 can now identify a protein’s three-dimensional structures from its amino-acid sequence to the width of an atom.
To give some context, AlphaFold2 competed with over 100 research groups worldwide in a competition known as the Critical Assessment of Protein Structure Prediction, or CASP. The goal was exactly what AlphaFold 2 achieved, to be able to predict a protein’s structure from its amino-acid sequence.
Evaluation
The main metric used to evaluate the accuracy of predictions was the Global Distance Test (GDT). It represents the percentage of amino acid residues within a threshold distance from the correct position. AlphaFold 2 achieved a median score of 87 GDT in the free-modeling category, up from AlphaFold’s initial score of just under 60 GDT in 2018.
Check out DeepMind’s full blog if you want to learn about their approach to solving the problem.
What does this mean for the future?
There are a number of positive consequences that can arise from this discovery. DeepMind has mentioned that it has opened opportunities in areas like biological research, drug design, and environmental sustainability. Andrei Lupas, an evolutionary biologist at the Max Planck Institute for Developmental Biology, said that "this will change medicine. It will change research. It will change bioengineering. It will change everything."
More tangibly, DeepMind sees this having a significant impact on drug development and finding treatments and vaccines faster.
Thanks for Reading!
This is huge news for Biology and Artificial Intelligence. It’s exciting and promising to see that Artificial Intelligence is having such a positive and large impact on science. If you want to read more about it, I’ve left some additional resources that you can refer to below:
AlphaFold: a solution to a 50-year-old grand challenge in biology
DeepMind’s protein-folding AI has solved a 50-year-old grand challenge of biology
Terence Shin
- If you enjoyed this, follow me on Medium for more
- Sign up for my email list here!
- Let’s connect on LinkedIn
- Interested in collaborating? Check out my website.