Proteins are essential to life, supporting practically all its functions. Yet, the structure of these complex molecules has stood as a grand challenge in biology for the past 50 years. However, with the recent developments in the field of artificial intelligence, it has become possible to predict the three-dimensional structure of how amino acids fold into proteins. A process that used to take 2 years, now takes 10 seconds.
John Jumper is a chemist, an expert on protein dynamics and senior researcher at DeepMind in London, where he and his team developed the AI system AlphaFold, which is capable of generating predictions of protein structures to atomic precision. At Falling Walls, Jumper discusses how he and his team were able to harness the power of artificial intelligence to make predictions about the shapes of proteins approximately 6 million times faster than was previously possible. This knowledge applied will help us understand how diseases progress, it will help us develop new drugs and unlock avenues of discovery that are currently blocked-off.