In this work we introduce a new method for protein coding regions identification with TBP, based on a newly transform, called Modified Morlet Transform (MMT), which does not need to be trained on sequences databases. We use a fixed binary mapping rules to create four binary sequences. Where each one represents the positions of each nitrogenate base in DNA sequence. Next the MMT, with different scales is applied to all binary sequences. The module of each normalized coefficient is projected onto the position axis. Projection onto the scale axis reveal which scale carry more signal energy throughout the positions. The result of the projection position axis represents the protein coding region identificator. These projection coefficients correspond to regions with TBP. Thus, we use thresholding coefficients, to exclude positions where the associated energy is lower.

The performance of the proposed transform was examined by analyzing synthetic and real DNA sequences. Preliminary results show that MMT is better than traditional methods by presenting greater sensitivity to TBP and discriminatory capability between protein coding regions.

This work is a principal part of my master's thesis.