Shannon's uncertainty principle and gene expression levels

Abstract

Shannon's uncertainty principle has been applied to measure the degree of constraints in codon bias in the coding sequences of Escherichia coli, Saccharomyces cerevisiae and Haemophilus influenzae. Our study shows a high degree of correlation between Shannon's uncertainty values and codon adaptation index (CAI). This result suggests that the degree of constraints determined by Shannon's uncertainty principle reflects the level of gene expression and we propose that Shannon's uncertainty principle can be used as an alternative method for predicting the level of gene expression. The main advantage of Shannon's uncertainty values over CAI is that for calculating Shannon's uncertainty values, one does not require any reference set of genes as required in CAI. This gives a potential use of Shannon's uncertainty principle over CAI in predicting the level of gene expression, specially for the newly sequenced genomes where genes are not properly annotated.