Hyperdimensional computing: A fast, robust, and interpretable paradigm for biological data

Area of Science:

• Bioinformatics
• Computational Biology
• Machine Learning

Background:

• Bioinformatics advances rely on algorithms for biological data analysis.
• Deep learning transformed sequence, structure, and functional analyses but is data-hungry and complex.
• Hyperdimensional computing (HDC) presents a novel, efficient alternative.

Purpose of the Study:

• To review and explore the potential of Hyperdimensional Computing (HDC) in bioinformatics.
• To highlight HDC's advantages over traditional deep learning methods.
• To assess HDC's applicability in diverse omics and health-related data analyses.

Main Methods:

• Exploration of HDC principles, representing biological concepts with high-dimensional random vectors.
• Analysis of HDC's unique operators for learning, reasoning, and querying.
• Review of existing literature and potential applications of HDC in bioinformatics.

Main Results:

• HDC demonstrates efficiency and interpretability in handling biological data.
• HD C excels with multimodal and structured data, overcoming deep learning limitations.
• HD C's vector-based approach offers a unique paradigm for biological data representation and analysis.

Conclusions:

• HDC shows significant promise as a powerful and accessible tool for modern bioinformatics.
• The efficiency and interpretability of HDC make it suitable for omics data searching and biosignal analysis.
• HDC applications extend to health informatics, offering new avenues for biological data interpretation.