Promoter sequences classification through the analysis of enthalpy, entropy, stability abd base-pair stacking values

Abstract

Promoter sequence recognition by RNAp enzyme is a key step in gene transcription. Its location is found in a few base pairs before the coding region. An in-depth study of promoters sequences role might provide an enhanced foundation to understand how genes are expressed under different conditions and produce biological rules to be used in computer techniques such as data clustering. Somehow, a cell behaves similarly as manmade machine, thus, its processes involve the best possible use of energy sources without producing too much heat. By this means, some physical concepts applied to machines, might, as well, be applied in cells, such as entropy and enthalpy variation. The present dissertation looks to assess the role of physical properties of the DNA: entropy, enthalpy, base-pair stacking and stability, in the characterization of Escherichia coli (E. coli) promoter sequences. To do so, a clustering technique was used to group promoter sequences clusters including the beforementioned features. With the cluster results in hand, a profile of the physical aspects of the DNA in promoter sequences may be drawn and biological inferences made upon these. Currently, not a big number of promoter identification tools make the use of combined profiles of enthalpy, entropy, base-pair stacking and stability. This paper has reported a strong correlation between enthalpy, stability and base pair stacking, where each combination of these features behaves differently in promoter sequences recognized by different sigma factors. We understand, according to the literature, that promoter sequences are known to be different in comparison to other genomic sequences, the results displayed in this paper enable a wider comprehension of difference between promoters themselves. Where, according to the sigma factor that is associated to the RNA polymerase recognition, the physical profile tends to be different, and by this, this paper’s results might bring a big acquisition to bioinformatics.