Machine learning
AI-driven antibody discovery and development: challenges and potential
Discovering and developing antibodies is extremely difficult, due at least in part to the extreme complexity of the antibody design space. Many are now turning to AI to help at various stages of this process.
Alchemab Therapeutics is a company with a clear understanding of both the potential and difficulties of the use of AI in the therapeutic antibody discovery space – an understanding that has led them to do things differently.
Find out more about how Alchemab are doing things differently here.
AI in bispecific antibodies and ADCs
Considering more specifically ADCs and bsAbs, AI has potential to assist in their development in a number of ways.
The first is in target identification and selection. Identifying single targets with therapeutic potential is a complex enough problem: identifying two targets with synergistic effect even more so. This is an essential challenge because developing and testing therapeutics is extremely costly, such that in the absence of good strategies to select promising targets, the search space is essentially limited to well-known targets with known antibodies.
The role of AI Patents in a Commercial Strategy
Whenever software is involved, the question of whether patents have value is going to be on the table. This is because historically software has had a reputation as ‘hard to patent’ and ‘hard to enforce’. While the former is not necessarily deserved, the latter is a valid point. However, there is value in AI patents in this field far beyond enforcement.
© 2025 Mewburn Ellis
