Zero-shot design of drug-binding proteins using neural networks

Add to your list(s) Download to your calendar using vCal

• Nicholas Polizzi
• Monday 20 October 2025, 12:30-13:30
• CRUK CI Lecture Theatre.

If you have a question about this talk, please contact Simona Valeviciute.

The use of deep neural networks has improved almost every aspect of protein structure prediction and design. Yet, the design of proteins that bind to small molecules has remained an outstanding challenge. Here, I will discuss new models and algorithms that unlock the use of neural networks for design of drug-binding proteins from scratch with unprecedented success rates. This work enables the design of binders in a single shot, which we demonstrate with the design of an exatecan-binding protein for applications in targeted delivery. We further devised an approach to turn this binder into an exatecan sensor. The de novo design of biosensors is an outstanding challenge in the field. I will finish my presentation by discussing our latest results that could potentially generalize de novo biosensor design.

This talk is part of the Seminars on Quantitative Biology @ CRUK Cambridge Institute series.

This talk is included in these lists:

• All CMS Events
• All Talks (aka the CURE list)
• Biology
• CCC talks for website
• CRUK CI Lecture Theatre
• CRUK CI Seminars
• CamBridgeSens
• Cambridge Centre for Data-Driven Discovery (C2D3)
• Cambridge talks
• Chris Davis' list
• Computational and Systems Biology
• Graduate-Seminars
• Interested Talks
• Liam
• Life Science Interface Seminars
• Life Sciences
• Life Sciences
• ME Seminar
• PMRFPS's
• School of Physical Sciences
• Seminars on Quantitative Biology @ CRUK Cambridge Institute
• Trust & Technology Initiative - interesting events
• bld31
• custom
• my_list
• ndk22's list
• ob366-ai4er
• other talks
• rp587
• se393's list
• sfm36
• yk449

Note that ex-directory lists are not shown.