what-is-what

As I used 2-track network to predict homodimer, the prediction map just shows the distance map of the monomer. Also, using complex modeling, two proteins are just predicted as an...

Variaous length of inverse folding sequences

1

We are currently using inverse folding in ESM3. When we increase the temperature from moderate 0.5 to 2 or 3, we see a wide range of sequence variation, which is...

How can we get attention weights from example sequence and structure?

2

How can we get attention weights from example sequence and structure? There were no arguments to get attention weights in transformer blocks, unlike esm2.