Artificial Intelligence for Design and Optimization of Antibodies for Targeted Therapies challenge
What is this challenge all about?
Antibody discovery and development has two main stages. The first one is to discover antibodies that bind the target of interest. The second, is to optimize the antibody to make a better drug. This challenge is about the second step. The starting point is a functional antibody that modulate the desired target. You are requested to propose and build a computational platform that delivers the sequence of a modified antibody sequence. Modifications are required at multiple levels:
• Manufacturing: ensure smooth antibody production at scale. Antibodies differ in their level
of solubility, expression levels in manufacturing cells, aggregation during purification, proper folding and post translational modification and more. A review that provides a detailed description can be found here.
• Physiological properties: pharmacokinetic and pharmacodynamics. Those terms relate to the concentration of the antibody in different tissues in the body and the rate of clearance from the body. A review article describing such parameters can be found here.
The above list of properties required is not complete. The goal is not to optimize for a specific antibody property but rather to suggest a process to optimize to any property.