After AlphaFold: Drug AI Targets PXR Activation
Get the Health newsletter
Daily health & science — research, biotech, public health, the studies worth knowing. Free.
- AlphaFold vaulted to prominence via the CASP competition in 2020 and went on to win a Nobel Prize
- Drug developers are now asking AI to solve practical problems like predicting whether the body will attack a drug candidate and render it useless
- PXR (pregnane X receptor) increases production of an enzyme that specifically breaks down foreign organic molecules, including drug molecules, so the body can dispose of them
- The enzyme PXR regulates metabolizes approximately 50% of all marketed drugs
- Most drug development campaigns only learn whether candidates activate PXR late in the game, forcing teams back to the drawing board
- An AI model that reliably predicts PXR activation addresses problems the source names outright — including drugs exiting the body too fast and drug-drug interactions
Why it matters: Roughly 50% of marketed drugs pass through the enzyme PXR regulates, and most campaigns only learn about PXR activation late — forcing teams back to the drawing board. Earlier prediction addresses two concrete problems the source names: drugs exiting the body too fast, and drug-drug interactions.
