Eur. Phys. J. B 47, 431-436 (2005)
https://doi.org/10.1140/epjb/e2005-00333-x

Size dependent complexity of sequences in protein families

¹ National laboratory of Solid State Microstructure, Institute of Biophysics, and Department of Physics, Nanjing University, 210093, P.R. China
² Interdisciplinary Center of Theoretical Studies, Chinese Academy of Sciences, Beijing 100080, P.R. China

Corresponding author: ^a wangwei@nju.edu.cn

Received: 11 January 2005
Revised: 6 June 2005
Published online: 28 October 2005

Abstract

The size dependent complexity of protein sequences in various families in the FSSP database is characterized by sequence entropy, sequence similarity and sequence identity. As the average length L_f of sequences in the family increases, an increasing trend of the sequence entropy and a decreasing trend of the sequence similarity and sequence identity are found. As L_f increases beyond 250, a saturation of the sequence entropy, the sequence similarity and the sequence identity is observed. Such a saturated behavior of complexity is attributed to the saturation of the probability P_g of global (long-range) interactions in protein structures when . It is also found that the alphabet size of residue types describing the sequence diversity depends on the value of L_f, and becomes saturated at 12.