Abstract
The immune system of vertebrates is generally viewed as a prototype of a highly adaptive, distributed, detection system, that identifies and neutralizes pathogenic intrusions. One of its puzzling features is that the immune receptors (antibodies) are able to bind to pathogens that they have not been `trained' to recognize. This anticipatory capability is thought to be due to a broad coverage of the pathogen space realized by the antibodies that the immune system can produce. What we would like to understand is how this coverage is achieved, given that the immune system uses a relatively small number of genes to construct its receptors. We use an evolutionary algorithm to explore the strategies that the antibody libraries may evolve in order to encode pathogen sets of various sizes. We derive a lower and an upper bound on the performance of the evolved antibody libraries as a function of their size and the length of the pathogen string. We also provide some insights in the strategy of the antibody libraries. We discuss the implications of our results for biological evolution of antibody libraries.
| Original language | English (US) |
|---|---|
| Pages (from-to) | 3793-3798 |
| Number of pages | 6 |
| Journal | Proceedings of the IEEE International Conference on Systems, Man and Cybernetics |
| Volume | 4 |
| State | Published - Dec 1 1998 |
| Externally published | Yes |
| Event | Proceedings of the 1998 IEEE International Conference on Systems, Man, and Cybernetics. Part 3 (of 5) - San Diego, CA, USA Duration: Oct 11 1998 → Oct 14 1998 |
ASJC Scopus subject areas
- Control and Systems Engineering
- Hardware and Architecture