We introduce a 4D numerical representation of DNA sequences and its application. Advantages and limitations are discussed. The method is illustrated by constructing sequence invariants on the coding sequence of the first exon of human β-globin gene. As application, we use the way, which is based on the construction of a vector whose components are the normalized leading eigen-values of the L/L matrices associated with DNA sequences, to examine the similarities/dissimilarities among the coding sequences of the first exon of P-globin gene of 11 different species. And state the easy way in counting the numbers of nucleotide substitutions by coordinates. © 2005 Elsevier B.V. All rights reserved.