Recognizing promoters in DNA using Bayesian neural networks

摘要

Binary data classification is to recognize positive data from unlabeled test data that may contain both positive and negative data. In this paper we propose a two-level approach to recognize E. Coli promoters in unlabeled DNA containing both promoter and non-promoter sequences. The first-level classifiers include three Bayesian neural networks which learn from three different feature sets. The outputs of the first-level classifiers are combined in the second-level to give the final result. Empirical study shows an excellent performance of the proposed approach.