Entropies of biosequences: The role of repeats

Abstract

DNA sequences of higher organisms contain thousands of nearly identical dispersed repetitive sequences. In order to understand the effect of such repeats on word entropies, we construct a model that can be analyzed analytically. The hypothetical model sequences consist of independent equidistributed symbols with randomly interspersed repeats. As a conclusion, we predict that the entropy of DNA sequences measuring the information content is much lower than suggested by earlier empirical studies. © 1994 The American Physical Society.