Assessing the Compositional Heterogeneity of Genomes

The chi-squared test of homogeneity is used to formulate a gross measure of compositional heterogeneity that is scale-invariant and independent of sequence origin. When computed from the position-dependent mononucleotide compositions of 24 microbial chromosome sequences, this (base 10) logarithmic measure varies from 1.3 to 3.6, the largest values found in sequences that are clearly separated into replichores by their skew diagrams. This heterogeneity measure is readily adapted to linear functions of the base fractions such as the G+C proportion used in identifying isochores