Information Technology Reference
In-Depth Information
Infection
(PAMPs)
Dangerous
a
c
d
e
f
b
Self
Nonself
abcdef
−
−
−
+
−
+
+
+
SNS
INS
DT
Responses to
each set
predicted by
−−
−
+
+
−
+
+
+
−
Figure 2.7
An abstract view of the antigen universe based on SNS, INS, and
DT models.
SNS model, the self-set consists of subsets
a
and
c
; thus, self-elements in
a
and
c
are
labeled as “
”. Howe ver,
according to the infectious nonself model (INS), only part of nonself is considered
infectious; thereby, nonself subsets
e
and
f
are labeled as “
−
”, whereas nonself subsets
b
,
d
,
e
, and
f
are labeled as “
+
”. However, in DT
(Matzinger, 2002), a subset of self is also considered to trigger alarm signals. h ere-
fore, self-subset
c
and nonself subsets
d
and
e
are labeled as “
+
+
”.
2.6 Computational Aspects of the Immune System
From the point of view of information processing, the biological immune system
exhibits many interesting characteristics; some of which are (Dasgupta, 1999)
•
Pattern matching
. h e immune system is able to recognize specifi c antigens
and generate appropriate responses. h is is accomplished by a recognition
mechanism based on chemical binding of receptors and antigens. h is bind-
ing depends on their molecular shapes and electrostatic charges.
•
Feature extraction
. Generally, immune receptors do not bind to a complete
antigen, but rather to portions of it (peptides). Accordingly, the immune sys-
tem can recognize an antigen just by matching segments of it. Peptides are
Search WWH ::
Custom Search