Artificial immune system

Artificial immune systems (AIS) are computational systems inspired by the principles and processes of the vertebrate immune system. The algorithms typically exploit the immune system's characteristics of learning and memory to solve a problem. They are coupled to artificial intelligence and some AIS algorithms (but not all) are closely related to genetic algorithms. More information on AIS can be found at *http://www.artificial-immune-systems.org/

Processes simulated in AlS include pattern recognition, hypermutation and clonal selection for B cells, negative selection of T cells, affinity maturation and immune network theory.

This article covers the algorithmic implementation of these processes. For underlying biological terminology, refer to the natural immune system.

Additional recommended knowledge

How to quickly check pipettes?

Guide to balance cleaning: 8 simple steps

What is the Correct Way to Check Repeatability in Balances?

Pattern recognition

Antibody & antigen representation is commonly implemented by strings of attributes. Attributes may be binary, integer or real-valued, although in principle any ordinal attribute could be used. Matching is done on the grounds of a distance metric, e.g. Euclidean distance, Manhattan distance or Hamming distance.

Hypermutation

Clonal selection algorithms are commonly used for antibody hypermutation. This allows the attribute string to be improved (as measured by a fitness function) using mutation alone. However, researchers argue that this clonal selection algorithm is similar to the mutation-based genetic algorithm and evolutionary strategies.

History

AIS began in the mid 70's with Farmer, Packard and Perelson's (1986) and Bersini and Varela's papers on immune networks (1990). However, it was only in the mid-90's that AIS became a subject area in its own right. Forrest et al (on negative selection) began in 1994; and Dasgupta conducted extensive studies on Negative Selection Algorithms. Hunt and Cooke started the works on Immune Network models in 1995; Timmis and Neal continued this work and made some improvements. De Castro & Von Zuben's and Nicosia & Cutello's work (on clonal selection) became notable in 2002. The first book on Artificial Immune Systems was edited by Dasgupta in 1999.

New ideas, such as danger theory and algorithms inspired by the innate immune system, are also now being explored. Although some doubt that they are yet offering anything over and above existing AIS algorithms, this is hotly debated, and the debate is providing one the main driving forces for AIS development at the moment.

Originally AIS set out to find efficient abstrations of processes found in the immune system but, more recently, it is becoming interested in modelling the biological processes and in applying immune algorithms to bioinformatics problems.

References

• J.D. Farmer, N. Packard and A. Perelson, (1986) "The immune system, adaptation and machine learning", Physica D, vol. 2, pp. 187--204
• H. Bersini, F.J. Varela, Hints for adaptive problem solving gleaned from immune networks. Parallel Problem Solving from Nature, First Workshop PPSW 1, Dortmund, FRG, October, 1990.
• D. Dasgupta (Editor), Artificial Immune Systems and Their Applications, Springer-Verlag, Inc. Berlin, January 1999, ISBN 3-540-64390-7
• L. DeCastro and J. Timmis (2001) "Artificial Immune Systems: A New Computational Intelligence Approach" ISBN 1-85233-594-7
• J Timmis, M Neal and J Hunt, (2000) "An Artificial Immune System for Data Analysis" pp. 143--150, Biosystems, no. 1/3, vol. 55.
• V. Cutello and G. Nicosia (2002) "An Immunological Approach to Combinatorial Optimization Problems" Lecture Notes in Computer Science, Springer vol. 2527, pp. 361-370.
• L. N. de Castro and F. J. Von Zuben, (1999) "Artificial Immune Systems: Part I -Basic Theory and Applications", School of Computing and Electrical Engineering, State University of Campinas, Brazil, No. DCA-RT 01/99.
• S. Garrett (2005) "How Do We Evaluate Artificial Immune Systems?" Evolutionary Computation, vol. 13, no. 2, pp. 145--178. http://mitpress.mit.edu/journals/pdf/EVCO_13_2_145_0.pdf
• V. Cutello, G. Nicosia, M. Pavone, J. Timmis (2007) An Immune Algorithm for Protein Structure Prediction on Lattice Models, IEEE Transactions on Evolutionary Computation, vol. 11, no. 1, pp. 101-117. http://www.dmi.unict.it/nicosia/papers/journals/Nicosia-IEEE-TEVC07.pdf