New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Option for sequence bias #3

Open

amin-sagar opened this issue Jan 20, 2025 · 1 comment

amin-sagar commented Jan 20, 2025

Hello.

I am wondering if it is possible to design peptides around known motifs. For example, if I know my peptide needs to have LXXXLXXL motif, can I add a sequence bias to drive the binders to have this motif?

I would be grateful for any suggestions.

Best,
Amin.

Collaborator

kxz18 commented Jan 21, 2025 •

edited

Loading

Though I didn't implement this, I think there might be two possible workarounds.

When decoding the residue types (here), force the desired position to be L.
Implementing guidance on the latent variable. You need to first derive the average latent embedding of L, then the guidance objective can be the distance between it and the generated latent variable at the desired positions.

Feel free to ask any other questions!

Best,
Xiangzhe.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment