PhD defence by Magnus Haraldson Høie

On Monday, 23 September, Magnus Haraldson Høie will defend his PhD thesis "AI in antibody design".

Time: 13:30
Place: Building 421, auditorium 74 & zoom: https://dtudk.zoom.us/meeting/register/u5Ysf-urqz0iEtFLOXtLBu--dy5MG7MBQbMR
Please be aware that the PhD defense may be recorded - This will also be informed at the beginning of the PhD defense.

Supervisor: Professor Morten Nielsen
Co-supervisor: Associate Professor Paolo Marcatili, Head of Antibody Design, Novo Nordisk
Professor Ole Winther, DTU Compute

Assessment committee:
Professor Ole Lund, DTU Health Tech
Professor Wouter Boomsma, University of Copenhagen
Head of BioAI Nicolas Lopez Carranza, InstaDeep

Chairperson:
Associate Professor Henrik Nielsen, DTU Health Tech

Abstract:
Proteins are essential molecules for life, performing virtually all biological processes in our cells. Understanding their function is key to many scientific advances, including the development of new protein drugs, such as antibodies used in immunotherapy.

This thesis introduces three new deep learning tools in protein design:

1. NetSurfP-3.0: Significantly speeds up protein secondary structure prediction using features from a protein language model, achieving a 700-fold increase in speed without losing accuracy.
2. DiscoTope-3.0: Enhances the prediction of antibody binding sites (epitopes) using a large protein structure-based deep learning model, improving accuracy for both experimental and predicted protein structures.
3. AntiFold: Improves antibody design by fine-tuning a large protein-structure based model, accurately designing antibody binding regions while maintaining their 3D structure.

By making these tools freely available online, this thesis aims to democratize access to advanced protein design techniques, paving the way for breakthroughs in medicine and biotechnology.