Recent investigations into global characteristics of long DNA sequences.

The sequencing of very large genes comprising tens and hundreds of thousands of bases raises important questions of how the data are to be viewed and analysed and what new information and rules on genome characterisation and organisation may exist in nature. Representation of such gene sequences require more compact techniques than the widely used letter-series method; several graphical techniques have been proposed that partially meet this need and reveal new patterns in sequence composition in conserved genes. Investigations using Chaos techniques have displayed self-similarities in the graphical representation of several classes of gene sequences that point to a fractal nature embedded in the sequence organisation in a scale-independent manner. Scale invariance is also seen in studies that have revealed unexpected correlations in gene sequences that imply that a nucleotide at any position influences nucleotides thousands of base positions away. These observations have given rise to a new way of looking at long gene sequences, but questions on the origin of such patterns and correlations and their implications in gene sequences and evolution remain to be resolved. This review provides a brief overview of the recent papers covering the techniques and the issues raised in these investigations on the global characteristics of long DNA sequences.