This comprehensive work deals with the applications of artificial intelligence in engineering, addressing both the theoretical and practical aspects of developing intelligent systems. Particular emphasis is placed on knowledge-based expert systems and on neural networks, with sections on knowledge representation, searching and solving strategies, blackboard systems, analogical, case-based and qualitative reasoning, and project management. Knowledge-based system building tools are discussed, along with practical tips on building large intelligent systems for solving engineering problems. Clearly explaining the key concepts with relevant examples, and with additional contributions from leading researchers in the field, this is a book written by an engineer for an engineering audience. Prepared with the graduate student in mind, it will also be of value as a general reference work in this highly topical area.

A high-affinity inhibitor protein called CIP, which can be produced by small truncations of p35, was earlier identified by Amin, Albers, and Pant. P35 is one of the physiological activators of cdk5, a member of the cyclin-dependent kinase family. These proteins are known to be associated with the hyperphosphorylation of specific neuronal proteins. This typically occurs in the case of neurodegenerative diseases such as Alzheimer s. In this paper the authors study in silico the binding mechanism of cdk5-p25 and cdk5-CIP complexes more in detail. This provides a better understanding of the inhibitory activity of the protein CIP. The authors use a geometry-based technique to verify the following hypothesis: p25 s truncation provides increased flexibility to CIP, and hence CIP is able to conform better to cdk5 interface than p25 is. Therefore CIP is expected to bind to cdk5 more easily than p25 and prevent it from reaching its active conformation. The authors' in silico study is based on a geometry-based alignment algorithm. The algorithm is capable of efficiently aligning two protein conformations with respect to their interfaces, which are represented as point sets. The algorithm is based on biochemical criteria as well as geometrical ones.