Artificial Immune Systems

A method for anomaly detection is introduced in which "normal" is defined by short-range correlations in a process' system calls. Initial experiments suggest that the definition is stable during normal behavior for standard UNIX programs. Further, it is able to detect several common intrusions involving sendmail and lpr. This work is part of a research program aimed at building computer security systems that incorporate the mechanisms and algorithms used by natural immune systems.

Intrusion detection based upon computational intelligence is currently attracting considerable interest from the research community. Characteristics of computational intelligence (CI) systems, such as adaptation, fault tolerance, high computational speed and error resilience in the face of noisy information, fit the requirements of building a good intrusion detection model. Here we want to provide an overview of the research progress in applying CI methods to the problem of intrusion detection. The scope of this review will encompass core methods of CI, including artificial neural networks, fuzzy systems, evolutionary computation, artificial immune systems, swarm intelligence, and soft computing. The research contributions in each field are systematically summarized and compared, allowing us to clearly define existing research challenges, and to highlight promising new research directions. The findings of this review should provide useful insights into the current IDS literature and be a good source for anyone who is interested in the application of CI approaches to IDSs or related fields.

D uring the last decade, the field of Artificial Immune System (AIS) is progressing slowly and steadily as a branch of Computational Intelligence (CI) as shown in .There has been increasing interest in the development of computational models inspired by several immunological principles. In particular, some are building models mimicking the mechanisms in the biological immune system (BIS) to better understand its natural processes and simulate its dynamical behavior in the presence of antigens/pathogens. Most of the AIS models, however, emphasize designing artifacts-computational algorithms, techniques using simplified models of various immunological processes and functionalities. Like other biologically-inspired techniques, such as artificial neural networks, genetic algorithms, and cellular automata, AISs also try to extract ideas from the BIS in order to develop computational tools for solving science and engineering problems. Although still relatively young, the Artificial Immune System (AIS) is emerging as an active and attractive field involving models, techniques and applications of greater diversity.

With increased global interconnectivity, reliance on e-commerce, network services, and Internet communication, computer security has become a necessity. Organizations must protect their systems from intrusion and computer-virus attacks. Such protection must detect anomalous patterns by exploiting known signatures while monitoring normal computer programs and network usage for abnormalities. Current antivirus and network intrusion detection (ID) solutions can become overwhelmed by the burden of capturing and classifying new viral stains and intrusion patterns. To overcome this problem, a self-adaptive distributed agent-based defense immune system based on biological strategies is developed within a hierarchical layered architecture. A prototype interactive system is designed, implemented in Java, and tested. The results validate the use of a distributed-agent biological-system approach toward the computer-security problems of virus elimination and ID.

After a decade of research into the area of artificial immune systems, it is worthwhile to take a step back and reflect on the contributions that the paradigm has brought to the application areas to which it has been applied. Undeniably, there have been a lot of successful stories-however, if the field is to advance in the future and really carve out its own distinctive niche, then it is necessary to be able to illustrate that there are clear benefits to be obtained by applying this paradigm rather than others. This paper attempts to take stock of the application areas that have been tackled in the past, and ask the difficult question ''was it worth it ?''. We then attempt to suggest a set of problem features that we believe will allow the true potential of the immunological system to be exploited in computational systems, and define a unique niche for AIS. #

Dendritic cells are antigen presenting cells that provide a vital link between the innate and adaptive immune system. Research into this family of cells has revealed that they perform the role of coordinating T-cell based immune responses, both reactive and for generating tolerance. We have derived an algorithm based on the functionality of these cells, and have used the signals and differentiation pathways to build a control mechanism for an artificial immune system. We present our algorithmic details in addition to some preliminary results, where the algorithm was applied for the purpose of anomaly detection. We hope that this algorithm will eventually become the key component within a large, distributed immune system, based on sound immunological concepts.

Biological systems such as human beings can be regarded as sophisticated information processing systems, and can be expected to provide inspiration for various ideas to science and engineering. Biologicallymotivated information processing systems can be classi ed into: brain-nervous systems (neural networks), genetic systems (evolutionary algorithms), and immune systems (arti cial immune systems). Among these, nervous systems and genetic systems have been widely applied to various elds. There have been a relative few applications of the immune system. This paper presents a survey of arti cial immune systems and provides numerous insights of immunity-based systems applications in science and engineering.

We investigated a real-valued Negative Selection Algorithm (NSA) for fault detection in man-in-the-loop aircraft operation. The detection algorithm uses body-axes angular rate sensory data exhibiting the normal flight behavior patterns, to generate probabilistically a set of fault detectors that can detect any abnormalities (including faults and damages) in the behavior pattern of the aircraft flight. We performed experiments with datasets (collected under normal and various simulated failure conditions) using the NASA Ames man-in-the-loop highfidelity C-17 flight simulator. The paper provides results of experiments with different datasets representing various failure conditions. Figure 1. The figure illustrates the concept of self and nonself in a feature space. Here F1, F2, etc. indicate different fault conditions represented by detectors.

Negative selection algorithms for hamming and real-valued shape-spaces are reviewed. Problems are identified with the use of these shape-spaces, and the negative selection algorithm in general, when applied to anomaly detection. A straightforward self detector classification principle is proposed and its classification performance is compared to a real-valued negative selection algorithm and to a one-class support vector machine. Earlier work suggests that realvalue negative selection requires a single class to learn from. The investigations presented in this paper reveal, however, that when applied to anomaly detection, the real-valued negative selection and self detector classification techniques require positive and negative examples to achieve a high classification accuracy. Whereas, one-class SVMs only require examples from a single class.

This paper presents a generalization of the graphbased genetic programming (GP) technique known as Cartesian genetic programming (CGP). We have extended CGP by utilizing automatic module acquisition, evolution, and reuse. To benchmark the new technique, we have tested it on: various digital circuit problems, two symbolic regression problems, the lawnmower problem, and the hierarchical if-and-only-if problem. The results show the new modular method evolves solutions quicker than the original nonmodular method, and the speedup is more pronounced on larger problems. Also, the new modular method performs favorably when compared with other GP methods. Analysis of the evolved modules shows they often produce recognizable functions. Prospects for further improvements to the method are discussed.

The (randomized) real-valued negative selection algorithm is an anomaly detection approach, inspired by the negative selection immune system principle. The algorithm was proposed to overcome scaling problems inherent in the hamming shape-space negative selection algorithm. In this paper, we investigate termination behavior of the realvalued negative selection algorithm with variable-sized detectors on an artificial data set. We then undertake an analysis and comparison of the classification performance on the high-dimensional KDD data set of the real-valued negative selection, a real-valued positive selection and statistical anomaly detection techniques. Results reveal that in terms of detection rate, real-valued negative selection with variable-sized detectors is not competitive to statistical anomaly detection techniques on the KDD data set. In addition, we suggest that the termination guarantee of the real-valued negative selection with variable-sized detectors is very sensitive to several parameters. 7 Received Operating Characteristic 8 generates detectors randomly

Accurate and fast islanding detection of distributed generation is highly important for its successful operation
in distribution networks. Up to now, various islanding detection technique based on communication,
passive, active and hybrid methods have been proposed. However, each technique suffers from
certain demerits that cause inaccuracies in islanding detection. Computational intelligence based techniques,
due to their robustness and flexibility in dealing with complex nonlinear systems, is an option
that might solve this problem. This paper aims to provide a comprehensive review of computational
intelligence based techniques applied for islanding detection of distributed generation. Moreover, the
paper compares the accuracies of computational intelligence based techniques over existing techniques
to provide a handful of information for industries and utility researchers to determine the best method
for their respective system.

In mobile ad-hoc networks, nodes act both as terminals and information relays, and they participate in a common routing protocol, such as Dynamic Source Routing (DSR). The networks are vulnerable to routing misbehavior, due to faulty or malicious nodes. Misbehavior detection systems aim at removing this vulnerability. For this purpose, we use an Artificial Immune System (AIS), a system inspired by the human immune system (HIS). Our goal is to build a system that, like its natural counterpart, automatically learns and detects new misbehavior. In this paper we build on our previous work [1,2] and investigate the use of four concepts: (1) “virtual thymus”, a novel concept, introduced in this paper, that provides a dynamic description of normal behavior in the system; (2) “clustering”, a decision making mechanism for decreasing false positive detections (3) “danger signal”, a concept that is, according to the “danger signal theory” of the human immune system [11,12], crucial for correct final decisions making; in our case, the signal is exchanged among nodes, which makes our detection system distributed; (4) “memory detectors”, used for achieving faster secondary response of the detection system. We implement our AIS in a network simulator and test it on two types of misbehavior. We analyze the performance and show the effects of the four concepts on the detection capabilities. In summary: thanks to the virtual thymus, the AIS does not require a preliminary learning phase in which misbehavior should be absent; the use of the clustering and the danger signal is useful for achieving low false positives; the use of memory detectors significantly accelerates the secondary response of the system.

This paper presents a comparative study of two important Clonal Selection Algorithms (CSAs): CLONALG and opt-IA. To deeply understand the performance of both algorithms, we deal with four different classes of problems: toy problems (one-counting and trap functions), pattern recognition, numerical optimization problems and NP-complete problem (the 2D HP model for protein structure prediction problem). Two possible versions of CLONALG have been implemented and tested. The experimental results show a global better performance of opt-IA with respect to CLONALG. Considering the results obtained, we can claim that CSAs represent a new class of Evolutionary Algorithms for effectively performing searching, learning and optimization tasks.

The Dendritic Cell algorithm (DCA) is inspired by recent work in innate immunity. In this paper a formal description of the DCA is given. The DCA is described in detail, and its use as an anomaly detector is illustrated within the context of computer security. A port scan detection task is performed to substantiate the influence of signal selection on the behaviour of the algorithm. Experimental results provide a comparison of differing input signal mappings.

Dendritic cells are antigen presenting cells that provide a vital link between the innate and adaptive immune system, providing the initial detection of pathogenic invaders. Research into this family of cells has revealed that they perform information fusion which directs immune responses. We have derived a Dendritic Cell Algorithm based on the functionality of these cells, by modelling the biological signals and differentiation pathways to build a control mechanism for an artificial immune system. We present algorithmic details in addition to experimental results, when the algorithm was applied to anomaly detection for the detection of port scans. The results show the Dendritic Cell Algorithm is successful at detecting port scans.

This paper addresses the problem of electric distribution network expansion under condition of uncertainty in the evolution of node loads in a time horizon. An immune-based evolutionary optimization algorithm is developed here, in order to find not only the optimal network, but also a set of suboptimal ones, for a given most probable scenario. A Monte-Carlo simulation of the future load conditions is performed, evaluating each such solution within a set of other possible scenarios. A dominance analysis is then performed in order to compare the candidate solutions, considering the objectives of: smaller infeasibility rate, smaller nominal cost, smaller mean cost and smaller fault cost. The design outcome is a network that has a satisfactory behavior under the considered scenarios. Simulation results show that the proposed approach leads to resulting networks that can be rather different from the networks that would be found via a conventional design procedure: reaching more robust performances under load evolution uncertainties.

Artificial immune systems, more specifically the negative selection algorithm, have previously been applied to intrusion detection. The aim of this research is to develop an intrusion detection system based on a novel concept in immunology, the Danger Theory. Dendritic Cells (DCs) are antigen presenting cells and key to the activation of the human immune system. DCs perform the vital role of combining signals from the host tissue and correlate these signals with proteins known as antigens. In algorithmic terms, individual DCs perform multi-sensor data fusion based on time-windows. The whole population of DCs asynchronously correlates the fused signals with a secondary data stream. The behaviour of human DCs is abstracted to form the DC Algorithm (DCA), which is implemented using an immune inspired framework, libtissue. This system is used to detect context switching for a basic machine learning dataset and to detect outgoing portscans in real-time. Experimental results show a significant difference between an outgoing portscan and normal traffic.

The concept of an information immune system (IIS) is introduced, in which undesirable information is eliminated before it can reach the user. The IIS is inspired by the natural immune systems that protect us from pathogens. IISs from multiple individuals can be combined to form a group IIS which filters out information undesirable to any of the members. The relationship between our proposed IIS architecture and the natural immune system is outlined, and potential applications, including information filtering, interactive design, and collaborative design, are discussed.

The immune system exhibits properties such as learning, distributivity continual adaptation, context dependent response and memory during the lifetime of a host. This paper argues that such properties are essential for the creation of future context-aware and ubiquitous systems where the need for such properties is becoming increasingly clear. To that end, we present an immune inspired system, which draws heavily on the immune network metaphor to create a metastable context-aware memory system that could be delivered in small hand-held devices.

In mobile ad-hoc networks, nodes act both as terminals and information relays, and participate in a common routing protocol, such as Dynamic Source Routing (DSR). The network is vulnerable to routing misbehavior, due to faulty or malicious nodes. Misbehavior detection systems aim at removing this vulnerability. In this paper we investigate the use of an Artificial Immune System (AIS) to detect node misbehavior in a mobile ad-hoc network using DSR. The system is inspired by the natural immune system of vertebrates. Our goal is to build a system that, like its natural counterpart, automatically learns and detects new misbehavior. We describe the first step of our design; it employs negative selection, an algorithm used by the natural immune system. We define how we map the natural immune system concepts such as self, antigen and antibody to a mobile ad-hoc network, and give the resulting algorithm for misbehavior detection. We implemented the system in the network simulator Glomosim; we present detection results and discuss how the system parameters impact the results. Further steps will extend the design by using an analogy to the innate system, danger signals, costimulation and memory cells.

This work presents the application of the omni-aiNet algorithm-an immune-inspired algorithm originally developed to solve single and multi-objective optimization problems-to the reconstruction of phylogenetic trees. The main goal here is to automatically evolve a population of phylogenetic unrooted trees, possibly with distinct topologies, by minimizing at the same time two optimization criteria: the minimum evolution and the mean-squared error. This proposal generates, in a single run, a set of nondominated solutions that represent the trade-offs of the two conflicting objectives, and gives the user the possibility of having distinct explanations for the differences observed at the terminal nodes of the trees. A series of experimental results is also reported in this work, in order to illustrate the effectiveness of the proposal and its capability to overcome the restrictive feedback provided by the application of well-known algorithms for phylogenetic reconstruction, such as the Neighbor Joining. Besides, the methodology presented in this work is compared to the popular NSGA-II algorithm, also modified to solve phylogenetic reconstruction problems.

Artificial immune systems have previously been applied to the problem of intrusion detection. The aim of this research is to develop an intrusion detection system based on the function of Dendritic Cells (DCs). DCs are antigen presenting cells and key to the activation of the human immune system, behaviour which has been abstracted to form the Dendritic Cell Algorithm (DCA). In algorithmic terms, individual DCs perform multi-sensor data fusion, asynchronously correlating the fused data signals with a secondary data stream. Aggregate output of a population of cells is analysed and forms the basis of an anomaly detection system. In this paper the DCA is applied to the detection of outgoing port scans using TCP SYN packets. Results show that detection can be achieved with the DCA, yet some false positives can be encountered when simultaneously scanning and using other network services. Suggestions are made for using adaptive signals to alleviate this uncovered problem.

The human immune system provides inspiration in the attempt of solving the knowledge acquisition bottleneck in developing ontology for semantic web application. In this paper, we proposed an extension to the Guided Agglomerative Hierarchical Clustering (GAHC) method that uses an Artificial Immune Network (AIN) algorithm to improve the process of automatically building and expanding the concept hierarchy. A small collection of Malay text is used from three different domains which are IT, Biochemistry and Fiqh to test the effectiveness of the proposed approach and also by comparing it with GAHC. The proposed approach consists of three stages: pre-processing, concept hierarchy induction using GAHC and concept hierarchy learning using AIN. To validate our approach, the automatically learned concept hierarchy is compared to a reference ontology developed by human experts. Thus it can be concluded that the proposed approach has greater ability to be used in learning concept hierarchy.

This work proposes a classification-rule discovery algorithm integrating artificial immune systems and fuzzy systems. The algorithm consists of two parts: a sequential covering procedure and a rule evolution procedure. Each antibody (candidate solution) corresponds to a classification rule. The classification of new examples (antigens) considers not only the fitness of a fuzzy rule based on the entire training set, but also the affinity between the rule and the new example. This affinity must be greater than a threshold in order for the fuzzy rule to be activated, and it is proposed an adaptive procedure for computing this threshold for each rule. This paper reports results for the proposed algorithm in several data sets. Results are analyzed with respect to both predictive accuracy and rule set simplicity, and are compared with C4.5rules, a very popular data mining algorithm.

In this paper, we have made medical application of a new artificial immune system named the information gain-based artificial immune recognition system (IG-AIRS) which is minimized the negative effects of taking into account all attributes in calculating Euclidean distance in shape-space representation which is used in many artificial immune systems. For medical data, microorganism dataset was applied in the performance analysis of our proposed system. Microorganism dataset was obtained using Cyranose 320 electronic nose. Our proposed system reached 92.35% classification accuracy with five-fold cross validation method. This result ensured that IG-AIRS would be helpful in classification of microorganism species based on laboratory tests, and would open the way to various microorganism species determine support by using electronic nose.

This work presents omni-aiNet, an immune-inspired algorithm developed to solve single and multi-objective optimization problems, either with single and multi-global solutions. The search engine is capable of automatically adapting the exploration of the search space according to the intrinsic demand of the optimization problem. This proposal unites the concepts of omni-optimization, already proposed in the literature, with distinctive procedures associated with immune-inspired concepts. Due to the immune inspiration, the omni-aiNet presents a population capable of adjusting its size during the execution of the algorithm, according to a predefined suppression threshold, and a new grid mechanism to control the spread of solutions in the objective space. The omni-aiNet was applied to several optimization problems and the obtained results are presented and analyzed.

Some biological phenomena offer clues to solving real-life, complex problems. Researchers have been studying techniques such as neural networks and genetic algorithms for computational intelligence and their applications to such complex problems. The problem of security management is one of the major concerns in the development of eBusiness services and networks. Recent incidents have shown that the perpetrators of cybercrimes are using increasingly sophisticated methods. Hence, it is necessary to investigate non-traditional mechanisms, such as biological techniques, to manage the security of evolving eBusiness networks and services. Towards this end, this paper investigates the use of an Artificial Immune System (AIS). The AIS emulates the mechanism of human immune systems that save human bodies from complex natural biological attacks. The paper discusses the use of AIS on one aspect of security management, viz. the detection of credit card fraud. The solution is illustrated with a case study on the management of frauds in credit card transactions, although this technique may be used in a range of security management applications in eBusiness.

Ensuring the security of computers is a nontrivial task, with many techniques used by malicious users to compromise these systems. In recent years a new threat has emerged in the form of networks of hijacked zombie machines used to perform complex distributed attacks such as denial of service and to obtain sensitive data such as password information. These zombie machines are said to be infected with a 'bot' -a malicious piece of software which is installed on a host machine and is controlled by a remote attacker, termed the 'botmaster of a botnet'. In this work, we use the biologically inspired Dendritic Cell Algorithm (DCA) to detect the existence of a single bot on a compromised host machine. The DCA is an immune-inspired algorithm based on an abstract model of the behaviour of the dendritic cells of the human body. The basis of anomaly detection performed by the DCA is facilitated using the correlation of behavioural attributes such as keylogging and packet flooding behaviour. The results of the application of the DCA to the detection of a single bot show that the algorithm is a successful technique for the detection of such malicious software without responding to normally running programs.

The immune system provides an ideal metaphor for anomaly detection in general and computer security in particular. Based on this idea, artificial immune systems have been used for a number of years for intrusion detection, unfortunately so far with little success. However, these previous systems were largely based on immunological theory from the 1970s and 1980s and over the last decade our understanding of immunological processes has vastly improved. In this paper we present two new immune inspired algorithms based on the latest immunological discoveries, such as the behaviour of Dendritic Cells. The resultant algorithms are applied to real world intrusion problems and show encouraging results. Overall, we believe there is a bright future for these next generation artificial immune algorithms.

"The Pareto front can provide valuable information on land-use planning decision by revealing the possible trade-offs among multiple, conflicting objectives. However, seeking the Pareto front of land-use allocation is much more difficult than finding a unique optimal solution, especially when dealing with large-area regions. This article proposes an improved artificial immune system for multi-objective land-use allocation (AIS-MOLA) to tackle this challenging task. The proposed AIS is equipped with three modified operators, namely (1) a heuristic hypermutation based on compromise programming, (2) a non-dominated neighbour-based proportional cloning and (3) a novel crossover operator that preserves connected patches. To validate the proposed algorithm, it was applied in a hypothetical land allocation problem. Compared with the Pareto Simulated Annealing (PSA) method, AIS-MOLA can generate solutions more approximate to the Pareto front, with computation time amounting to only 5.1% of PSA. In addition, AIS-MOLA was also applied in the case study of Panyu, Guangdong, PR China, a large area with 389 × 337 cells. Experimental results indicate that this algorithm, even dealing with large-area land-use allocation problems, is capable of generating optimal alternative solutions approximate to the true Pareto front.Moreover, the distribution of these solutions can quantitatively demonstrate the complex trade-offs between the spatial suitability and the compactness in the study area. Software and supplementary materials are available at http://www.geosimulation.cn/AIS-MOLA/."

In this article, we present a parallel image processing system based on the concept of reactive agents. Our system lies in the oRis language, which allows to describe ÿnely and simply the agents' behaviors to detect image features. We also present a method of segmentation using a multi-agent system, and two biological applications made with oRis. The stopping of this multi-agent system is implemented through a technique issued from immunology: the apoptosis.

A sensor network is a collection of wireless devices that are able to monitor physical or environmental conditions. These devices (nodes) are expected to operate autonomously, be battery powered and have very limited computational capabilities. This makes the task of protecting a sensor network against misbehavior or possible malfunction a challenging problem. In this document we discuss performance of Artificial immune systems (AIS) when used as the mechanism for detecting misbehavior.

This paper presents a framework to generate multi-shaped detectors with valued negative selection algorithms (NSA). In particular, detectors can take the form of hyper-rectangles, hyper-spheres and hyper-ellipses in the non-self space. These novel pattern detectors (in the complement space) are evolved using a genetic search (the structured genetic algorithm), which uses hierarchical genomic structures and a gene activation mechanism to encode multiple detector shapes. This genetic search (the structured GA) allows in maintaining diverse shapes while contributing to the proliferation of best suited detector shapes in expressed phenotype. The results showed that a significant coverage of the non-self space could be achieved with fewer detectors compared to other NSA approaches (using only single-shaped detectors). The uniform representation scheme and the evolutionary mechanism used in this work can serve as a baseline for further extension to use several shapes, providing an efficie...

We present a swarm-based, 3-dimensional model of the human immune system and its response to first and second viral antigen exposure. Our model utilizes a decentralized swarm approach with multiple agents acting independently—following local interaction rules—to exhibit complex emergent behaviours, which constitute externally observable and measurable immune reactions. The two main functional branches of the human immune system, humoral and cell-mediated immunity, are simulated. We model the production of antibodies in response to a viral population; antibody-antigen complexes are formed, which are removed by macrophages; virally infected cells are lysed by cytotoxic T cells. Our system also demonstrates reinforced reaction to a previously encountered pathogen, thus exhibiting realistic memory response.

Do artificial immune systems (AIS) have something to offer the world of optimisation? Indeed do they have any new to offer at all? This paper reports the initial findings of a comparison between two immune inspired algorithms and a hybrid genetic algorithm for function optimisation. This work is part of ongoing research which forms part of a larger project to

Using hyperspheres as antibody recognition regions is an established abstraction which was initially proposed by theoretical immunologists for use in the modeling of antibody-antigen interactions. This abstraction is also employed in the development of many artificial immune system algorithms. Here, we show several undesirable properties of hyperspheres, especially when operating in high dimensions and discuss the problems of hyperspheres as recognition regions and how they have affected overall performance of certain algorithms in the context of real-valued negative selection.

MHC molecules, also known in the human as human leukocyte antigens (HLA), display peptides on antigen presenting cell surfaces for subsequent T cell recognition. Identification of these antigenic peptides is especially important for developing peptide-based vaccines. Consequently experimental and computational approaches have been developed for their identification. A major impediment to such an approach is the extreme polymorphism of HLA, which is in fact the basis for differential peptide binding. This problem can be mitigated by the observation that despite such polymorphisms, HLA molecules bind overlapping set of peptides, and therefore, may be grouped accordingly into supertypes. Here we describe a method of grouping HLA alleles into supertypes based on analysis and subsequent clustering of their peptide binding repertoires. Combining this method with the known allele and haplotype gene frequencies of HLA I molecules for five major American ethnic groups (Black, Caucasian, Hispanic, Native American, and Asian), it is now feasible to identify supertypic combinations for prediction of antigenic peptide, offering the potential to generate peptide-vaccines with a population coverage ≥95%, regardless of ethnicity. One combination including five distinct supertypes is available online at our PEPVAC web server (http://immunax.dfci.harvard.edu/PEPVAC/). Promiscuous peptides predicted to bind to these five supertypes represent around 5% of all possible peptide binders from a given genome.

... that in a certain part of the body an immune response of one class may be efficient, but the ... We feel this would be particularly suitable for a dynamic environment such as the internet. ... Similarly a web usage mining system may be implemented with an e-commerce site in mind. ...

Many recent advances have been made in understanding the functional implications of the global topological properties of biological networks through the application of complex network theory, particularly in the area of small-world and scalefree topologies. Computational studies which attempt to understand the structure-function relationship usually proceed by defining a representation of cells and an affinity measure to describe their interactions. We show that this necessarily restricts the topology of the networks that can arise-furthermore, we show that although simple topologies can be produced via representation and affinity measures common in the literature, it is unclear how to select measures which result in complex topologies, for example, exhibiting scalefree functionality. In this paper, we introduce the concept of the potential network as a method in which abstract network topologies can be directly studied, bypassing any definition of shape-space and affinity function. We illustrate the benefit of the approach by studying the evolution of idiotypic networks on a selection of scale-free and regular topologies, finding that a key immunological property-tolerance-is promoted by bipartite and heterogeneous topologies. The approach, however, is applicable to the study of any network and thus has implications for both immunology and artificial immune systems.