Application of discrete Fourier inter-coefficient difference for assessing genetic sequence similarity

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Digital signal processing (DSP) is increasingly applied to biological sequence analysis due to the digital nature of genetic data.
• DSP methods have shown success in identifying coding regions and are now being used for DNA gene similarity.
• Traditional DNA comparison often relies on sequence alignment, which can be computationally intensive and may miss subtle similarities.

Purpose of the Study:

• To introduce the inter-coefficient difference (ICD) transformation, a novel DSP technique for DNA sequence analysis.
• To present ICD as an alignment-free method for establishing genetic signatures and measuring DNA sequence similarity.
• To evaluate the accuracy of the ICD method in phylogenetic tree construction compared to traditional alignment techniques.

Main Methods:

• The inter-coefficient difference (ICD) transformation, an extension of the discrete Fourier transformation, was applied to DNA sequences.
• The method generates a unique genetic signature for each DNA sequence.
• Phylogenetic trees were constructed using ICD-derived similarity measures and compared with trees from traditional alignment methods.

Main Results:

• The ICD transformation successfully generated genetic signatures for DNA sequences.
• The method demonstrated accurate relative measures of similarity among diverse DNA sequences, including insulin genes and avian influenza viruses.
• Phylogenetic trees generated using the ICD method showed high accuracy, comparable to traditional alignment-based approaches, without prior alignment.

Conclusions:

• The inter-coefficient difference (ICD) transformation is a robust, alignment-free method for DNA sequence comparison and phylogenetic analysis.
• This novel DSP technique offers an accurate and efficient alternative to traditional sequence alignment for establishing genetic similarity.
• The ICD method has broad applicability in genomics and bioinformatics for analyzing biological sequences across different species and viral strains.