Hadamard code based numerical mapping approach for the identification of protein coding regions in genomic sequences...

Area of Science:

• Biomedical Engineering
• Genomics
• Computational Biology

Background:

• Digital Signal Processing (DSP) is crucial for analyzing genomic sequences.
• Identifying protein-coding regions in DNA relies on detecting period-3 patterns.
• Existing numerical encoding methods for DNA analysis have limitations.

Purpose of the Study:

• To introduce a novel Hadamard Based Numerical Encoding (HBNE) method for improved protein-coding region prediction.
• To integrate HBNE with advanced signal processing techniques like Elliptic filters and Gaussian windowing.
• To enhance the accuracy of identifying coding regions in Deoxyribo Nucleic Acid (DNA) sequences.

Main Methods:

• Developed a Hadamard Based Numerical Encoding (HBNE) for DNA sequences.
• Applied signal processing techniques, including Elliptic filters with Gaussian windowing.
• Utilized Discrete Fourier Transform (DFT) for period-3 property detection and noise suppression.

Main Results:

• The HBNE method achieved 95% accuracy in predicting protein-coding regions.
• The Area Under the Curve (AUC) reached 92% for the proposed method.
• HBNE outperformed existing encoding methods like Voss, TBNE, FBNE, IBNE, and WCBNE.

Conclusions:

• The HBNE method offers a significant advancement in the accurate prediction of protein-coding regions.
• This approach effectively combines numerical encoding with DSP for enhanced genomic sequence analysis.
• HBNE provides a more robust tool for biomedical engineering and genetic research.