The next Shocklab Seminar will be held in-person (with an online component available) on Wednesday, 6 May 2026 from 16h00 - 17h00 (UTC+2).

Lood van Niekerk from the ILINA Program and the African Hub on AI Safety, Peace and Security will be presenting about Machine Learning for Protein Modeling and Design.

Title: Machine Learning for Protein Modeling and Design
Speaker: Lood van Niekerk
Date: Wednesday, 6 May 2026
Time: 16:00-17:00 (GMT +2)
Venue: Room M209, UCT Department of Mathematics and Applied Mathematics Mathematics Building, 7 University Ave N, Rondebosch, Cape Town, 7700 
Virtual Meeting Link:  https://bit.ly/shocklab-seminars

 
Abstract: 

Proteins are the machines that perform most of the function in our cells and in the cells of all organisms (replicating DNA, performing reactions, responding to stimuli). A better understanding of proteins can be used for designing medicines, genetically engineering organisms and designing highly efficient catalysts (enzymes). We'll discuss recent advances in protein sequence and structure models, applications, and current drawbacks in the field.

Bio:

Lood studied Computer Science at Stellenbosch with a focus on statistics / data science, then a Masters at Oxford. He then worked for 2 years as a research assistant in Debora Marks's Lab at Harvard, where he developed and benchmarked different protein sequence models. He then worked at Ginkgo Bioworks, a synthetic biology company, working on diffusion models for protein sequences and antibody property prediction. He is now working at a small startup developing generative models for designing new medicines (small molecules).

 

Housekeeping: