Self-organizing algorithm gene expression data

摘要

self-organizing for mass analysis of gene expression data from experiments DNA arrays, starting algorithm based on a composite of two neurons "daughters" connected to a neuron "mother" structure, which are composed of a list of values ​​or profile, initially of random data. The algorithm, in a succession of cycles, is dividing the data set into subsets. Each cycle consists of a number of times in which data terminals are presented and these neurons are updated. When between two consecutive periods updating produced is below a certain preset minimum level, it is considered that the network has converged in that cycle and it is decided whether the network should continue to grow or have already stopped growing, in which case it is given for completion of the process or algorithm.