Abstract:Introduction: Identifying and predicting protein-coding regions within DNA sequences play a pivotal role in genomic research. This paper introduces an approach for identifying proteincoding regions in DNA sequences, employing a hybrid methodology that combines a digital bandpass filter with wavelet transforms and various spectral estimation techniques to enhance exon prediction. Specifically, the Haar and Daubechies wavelet transforms are applied to improve the accuracy of protein-coding region (exon) prediction, enabling the extraction of intricate details that may be obscured in the original DNA sequences. background: The identification and prediction of protein-coding regions within DNA sequences play a pivotal role in genomic research. Methods: This research showcases the utility of Haar and Daubechies wavelet transforms, both nonparametric and parametric spectral estimation methods, and the deployment of a digital band pass filter for detecting peaks in exon regions. Additionally, the application of the Electron-Ion Interaction Potential (EIIP) method for converting symbolic DNA sequences into numerical values and the utilization of sum-of-sinusoids (SoS) mathematical models with optimized parameters further enrich the toolbox for DNA sequence analysis, ensuring the success of this proposed method in modeling DNA sequences optimally and accurately identifying genes. objective: Enhanced Protein-Coding Region Identification in DNA Sequences Using Wavelet Transforms Results: The outcomes of this approach showcase a substantial enhancement in identification accuracy for protein-coding regions. In terms of peak location detection, the application of Haar and Daubechies wavelet transforms enhances the accuracy of peak localization by approximately (0.01, 3-5 dB). When employing non-parametric and parametric spectral estimation techniques, there is an improvement in peak location by approximately (0.01, 4 dB) compared to the original signal. The proposed approach also achieves higher accuracy when compared with existing methods. method: hybrid methodology that combines a digital band-pass filter with wavelet transforms and various spectral estimation techniques to enhance exon prediction. Conclusion: These findings not only bridge gaps in DNA sequence analysis but also offer a promising pathway for advancing exonic region prediction and gene identification in genomics research. The hybrid methodology presented stands as a robust contribution to the evolving landscape of genomic analysis techniques. result: The results obtained through this proposed method demonstrate significantly improved identification accuracy. These findings offer a promising avenue for DNA sequence analysis, exonic region prediction, and gene identification.

Protein coding regions prediction by fusing DNA shape features

Gene Prediction by the Noise-Assisted MEMD and Wavelet Transform for Identifying the Protein Coding Regions

Predicting DNA structure using a deep learning method

Prediction of DNA-binding Residues from Sequence Information Using Convolutional Neural Network.

Research on DNA-Binding Protein Identification Method Based on LSTM-CNN Feature Fusion

Exon prediction based on multiscale products of a genomic-inspired multiscale bilateral filtering

NCNet: Deep Learning Network Models for Predicting Function of Non-coding DNA

An improved deep learning method for predicting DNA-binding proteins based on contextual features in amino acid sequences

Prediction of DNA binding proteins using local features and long-term dependencies with primary sequences based on deep learning

CNNsite: Prediction of DNA-binding Residues in Proteins Using Convolutional Neural Network with Sequence Features

Prediction of Protein Coding Regions by the 3-Base Periodicity Analysis of a DNA Sequence.

Coding Region Prediction Based on a Universal DNA Sequence Representation Method

Deep learning based DNA:RNA triplex forming potential prediction

ncRNA Coding Potential Prediction Using BiLSTM and Transformer Encoder-Based Model

Modeling in-vivo protein-DNA binding by combining multiple-instance learning with a hybrid deep neural network

Fusion Neural Network (FusNet) for predicting protein-mediated loops

DNA-Level Splice Junction Prediction using Deep Recurrent Neural Networks

Improved Hybrid Approach for Enhancing Protein Coding Regions Identification in DNA Sequences

A Novel Ncrna Gene Prediction Approach Based on Fuzzy Neural Networks with Structure Learning

DeepSite: bidirectional LSTM and CNN models for predicting DNA–protein binding

PDNAPred: Interpretable prediction of protein-DNA binding sites based on pre-trained protein language models