“…The recent reviews (Davies et al, 2008;Vinga and Almeida, 2003) on published methods of alignment-free sequence comparison report several concepts of distance measures, such as Markov chain models and Kullback-Leibler discrepancy (Wu et al, 2001), chaos theory (Almeida et al, 2001), Kolmogorov complexity , decision tree induction algorithm (Huang et al, 2004), graphical representation (Liao and Wang, 2004;Randic et al, 2003;Yau et al, 2003), probabilistic measure (Pham and Zuegg, 2004;Yu et al, 2011). Furthermore, sequence vector representation approaches without alignment are also prevalent, such as composition vector (Chan et al, 2010;Chu et al, 2004;Gao and Qi, 2007;Qi et al, 2004), feature vector (Carr et al, 2010;Liu et al, 2006), and moment vector Yu et al, 2010). However, all of these methods treat the DNA or protein sequences as simple character strings, and then consider occurrence frequencies and distribution of nucleotides or amino acids, or complexity of sequences and so on.…”