Articulation and Clarification of the Dendritic Cell Algorithm - Artificial Immune Systems - page 405

The aim of this paper is to demonstrate the anomaly detection capabilities of
the DCA and to clarify which features of the algorithm facilitate detection.

2 Dendritic Cells in vivo

The DCA is based on the function of dendritic cells whose primary role is as an
antigen presenting cell. DCs behave very differently to the cells of the adaptive
immune system. Before describing the function of the algorithm we give a general
overview of DC biology, introducing different cells, organs and their behaviour.
More information on natural DCs can be found in [9].
In vivo , DCs can perform a number of different functions, determined by their
state of maturation. Modulation between these states is facilitated by the de-
tection of signals within the tissue - namely danger signals, PAMPs (pathogenic
associated molecular patterns), apoptotic signals (safe signals) and inflamma-
tory cytokines which are described below. The maturation state of a DC is de-
termined by the relative concentrations of these four types of signal. The state
of maturity of a DC influences the response by T-cells, to either an immuno-
genic or tolerogenic state, for a specific antigen. Immature DCs reside in the
tissue where they collect antigenic material and are exposed to signals. Based on
the combinations of signals received, maturation of the DCs occurs generating
two terminal differentiation states, mature or semi-mature. Mature DCs have an
activating effect while semi-mature DCs have a suppressive effect. The different
output signals (termed output cytokines) generated by the two terminal states of
DCs differ suciently to provide two different contexts for antigen presentation,
shown abstractly in Figure 1.

Fig. 1. An abstract view of DC maturation and signals required for differentiation.
CKs denote cytokines