Proteins are extremely complex because their function is not just a function of the amino acids that they are built from, but also their geometry. Proteins can be very large molecules and they “fold” into complex shapes. Understanding their geometry is necessary in order to understand their function, and to develop medications. The process is so intensive that a distributed system was setup called Folding@Home which allows you to donate computer time to the global effort to catalog proteins. I’ve done this for many years.
A recent breakthrough from DeepMind, a subsidiary of Alphabet, has developed a second generation AI system that is more effective at predicting the shape of proteins than any known computational method. On a scale of 0-100 the AI, AlphaFold 2, achieves a score of 87. This compares favorably with experiments which score 90 and are very time and labor intensive.
This computational work represents a stunning advance on the protein-folding problem, a 50-year-old grand challenge in biology. It has occurred decades before many people in the field would have predicted. It will be exciting to see the many ways in which it will fundamentally change biological research.
PROFESSOR VENKI RAMAKRISHNAN
NOBEL LAUREATE AND PRESIDENT OF THE ROYAL SOCIETY
We are living in the future. These advances may seem trivial over the next five years but they will revolutionize medicine over the next three decades.