Statistical Methods for Detecting Latent Periodicity in Biological Sequences: Solving a Problem of Small-Size Samples

摘要

An original spectral-statistical approach for detecting latent periodicity in biological sequences is proposed. This approach can be applied under conditions of limited statistical sample. It allows one to avoid redundancy and instability when identifying the latent periodicity structure. The results of spectral-statistical approach application for latent periodicity search in genomes of model organisms are collected in the HETEROGENE database http://www.jcbi.ru/lp_baze/.