Bioinformatics data mining using artificial immune systems and neural networks

Abstract

-Bioinformatics is a data-intensive field of researchIn a negative selection algorithm, detectors are generated and development. The purpose of bioinformatics data mining is randomly first, then they are evolved (i.e., eliminated if they to discover the relationships and patterns in large databases to match any "self" samples) to obtain a set of trained "mature" provide useful information for biomedical analysis and diagnosis.detectors. In the testing mode, each unknown data instance is In this research, algorithms based on artificial immunepresented to the detector set and classified as either "self" or systems (AIS) and artificial neural networks (ANN) are employed "non-self". That is, if the unknown data instance matches any for bioinformatics data mining. Three different variations of the detector in the detector set, then it is classified as "non-self" or real-valued negative selection algorithm and a multi-layer feedforward neural network model are discussed, tested and an anomaly; while on the other hand, if the incoming data compared via computer simulations. It is shown that the ANN instance is not recognized by any detector, it is considered to model yields the best overall result while the AIS algorithm is be a member of the "self" set.