Inferring Helitron Structures from 1D and 2D Representations Based on the Chaos Game Theory

This work introduces a new procedure to highlight helitron structures in the C. elegans DNA sequences. In this sense, the DNA strings are converted to numerical representations either in one or two dimensions. As a one dimensional coding, we choose the Frequency Chaos Game Signal (FCGS) which encodes the DNA based on the Chaos Game theory. Afterwards, a time-frequency representation - generated by the complex Morlet wavelet analysis - transforms the obtained signal into a 2D representation. Through the examples furnished in this paper, we demonstrate the accuracy of the method in emphasizing the helitron borders especially with high orders of FCGS. A comparison with the 2D spectrogram representation confirms the obtained results.