Information Technology Reference
In-Depth Information
approach [7], which promotes the use of an interdisciplinary set of stages to
develop and analyse bio-inspired algorithms in a more principled way.
As a continuation of these ideas we present in this paper the initial stages of
our work aimed at exploiting the notion of degenerate immune receptors for use
in AIS. By following the conceptual framework approach [7], we have used bio-
logical detail drawn from the immunological research literature to build a com-
putational model based on the process of T
H
cell activation in the paracortex
of a lymph node, in which the T
H
cell receptors are assumed to be degenerate.
We believe that the model we have developed is a novel tool with a richness
of behaviour to enable the investigation of abstract degenerate detectors. Ulti-
mately this investigation aims to generate enough insight to extract algorithmic
design principles that will benefit the development of an AIS for pattern classi-
fication. It is noted that the degeneracy issues we explore here are not explicitly
connected to the ideas of immune networks.
2
Degeneracy
Degeneracy is a property that is not only seen in the immune system, but,
according to Edelman and Gally [8], is a ubiquitous biological property present
at most levels of biological organisation. They define degeneracy in biology as:
“the ability of elements that are structurally different to perform the
same function or yield the same output”
Examples they give include the genetic code, where different sequences can en-
code the same polypeptide, and human language, where there many different
ways to transmit the same message. They go on to argue that the omnipresence
of degeneracy in biology is a result of it being conserved and favoured by natural
selection. Additionally, it is noted that degeneracy in biological systems is typi-
cally accompanied with complexity, and it is suggested that degeneracy plays a
key role in complex systems.
Parnes [9] states that even though degeneracy is a term that has been used
in immunology for the last 35 years, it has escaped rigorous definition. For our
work, we have adopted the definition given by Cohen [6], which describes antigen
receptor degeneracy as the:
“capacity of any single antigen receptor to bind and respond to (recog-
nize) many different ligands”
Cohen [6] reports that the main consequence of the degeneracy of antigen recep-
tors is
poly-recognition
, whereby a single lymphocyte clone can recognise different
antigen epitopes. This causes a problem for the traditional clonal selection theory
view of immunology [10] that relies on the strict specificity of lymphocyte clones.
In [9], Parnes notes that in immunology there is a notable confusion between the
ideas of 'degeneracy', 'cross-reactivity' and 'promiscuity'. The interested reader
is referred to the Parnes [9] article for a detailed description of this issue.
