Statistical Prediction of Admissible Zones in a Sequence of Symbols from a Finite Alphabet: II.

摘要

Complex algorithms that make it possible to increase significantly (in comparison with the algorithms based on the method of the maximum of α posteriori probability) the prediction accuracy for the boundaries of coding and noncoding zones of a given sequence of nucleotides (boundaries of genes and "nongenes") are presented and studied on the test samples supplied by the GenBank database 1997.