Jyunmi

And you even have its 3D structure, which is like knowing the exact shape of that machine.

But every time you try to design a drug to stick to it, nothing works.

In pharma, people often call these undruggable targets.

That just means this protein really matters for the disease, but none of our usual drug tricks can hit it in a clean way.

A team at MIT and several partner labs just built an AI model aimed straight at that frustration.

It's called BoltsGen.

Its job is to design new protein binders for hard cards.

A binder is a molecule that latches onto a specific target in the body.

If you can get a binder that sticks tightly and only where you want, you have the starting point for many modern drugs like antibodies or engineered proteins.

So what does the AI do?

Boltzgen is a generative model.

That means it does not just label the data.

It creates new examples that look physically realistic.

You feed it the 3D structure of the target.

Think of that as giving the AI a detailed map of the protein surface.

You can also give it a few design hints like stay small or attach here.

In return, Boltzten proposes brand new protein shapes and amino acid sequences.

An amino acid sequence is just the letter string that tells a protein how to fold.

The model tries to pick sequences that will fold up into a stable shape and hug the target at the right spots.

So the system takes, here's the thing we want to hit, and answers with, here's some new molecules that look like they could hit it.