AI-designed Antibodies: Not another medical breakthrough story?

Oh no, not another medical breakthrough story on LSS? What happened to all those cocktails? What about all the economics and evolution stuff we used to get? Yet let us remind you of the words of our Founding Charter,* gentle readers: this started out as an antibiotics website, even more so when it was a post on Facebook. Of course when supplementary techniques like bacteriophages and improved antibodies came along, we welcomed those too. Which is why we have to give this new report from Nature Briefings AI helps design antibodies from scratch, front and centre on today’s blog. When you read this, you will understand why

For the first time, an artificial intelligence (AI) system has helped researchers to design completely new antibodies. Creating new versions of these immune proteins, which can be used as drugs, is usually a lengthy and costly process. An AI algorithm similar to those of the image-generating tools Midjourney and DALL·E was trained on thousands of real-world structures of antibodies attached to their target proteins. It then churned out thousands of new antibodies that recognize certain bacterial, viral or cancer-related targets. Although in laboratory tests only about one in 100 designs worked as hoped, biochemist and study co-author Joseph Watson says that “it feels like quite a landmark moment”.Nature | 4 min read
Reference: bioRxiv preprint (not peer reviewed)

Once again, we are at the start here. It’s still, at proof of concept stage, and awaits peer review. On the other hand, this is genuine landmark moment which people of the future will look back on. And whatever you are doing now as you read this, remember it. You were there on the day it happened.

• a copy of the founding charter is still visible in the flowerbed round the side near the car park, the bit where all the smokers go. Actually, could we clean it up a bit?