AI turns its artistry to creating new human proteins

“One of the most powerful things about this technology is that, like DALL-E, it does what you tell it to do,” said Nate Bennett, one of the researchers working at the University of Washington lab. “An endless number of designs can be generated from a single prompt.”

The Rise of OpenAI

The San Francisco-based company is one of the world’s most ambitious artificial intelligence laboratories. Here’s a look at some recent developments.

To generate images, DALL-E relies on what artificial intelligence researchers call a neural network, a mathematical system loosely modeled on the network of neurons in the brain. This is the same technology that recognizes the commands you bark into your smartphone, enables self-driving cars to identify (and avoid) pedestrians, and translates languages ​​on services like Skype.

A neural network learns skills by analyzing huge amounts of digital data. For example, by finding patterns in thousands of corgi photos, it can learn to recognize a corgi. Using DALL-E, researchers built a neural network that looked for patterns while analyzing millions of digital images and the text captions that described what each of those images represented. In this way, it learned to see the connections between the images and the words.

When you describe an image for DALL-E, a neural network generates a set of key features that image may contain. A feature could be the curvature of a teddy bear’s ear. Another might be the line on the edge of a skateboard. Then a second neural network — called the diffusion model — creates the pixels needed to realize those features.

The diffusion model is trained on a series of images in which noise – imperfections – is gradually added to a photo until it becomes a sea of ​​random pixels. As it analyzes these images, the model learns to perform this process in reverse. If you feed it random pixels, it removes the noise and converts those pixels into a coherent image.

At the University of Washington, other academic labs, and new startups, researchers are using similar techniques to make new proteins.

Proteins begin as chains of chemical bonds that then twist and fold into three-dimensional shapes that define their behavior. In recent years, artificial intelligence labs like DeepMind, owned by Alphabet, the same parent company as Google, have shown that neural networks can accurately guess the three-dimensional shape of any protein in the body based only on the smaller compounds it contains — at vaster scales scientific progress.