A novel lossless encoding algorithm for data compression-genomics data as an exemplar.

Area of Science:

• Bioinformatics
• Computer Science
• Data Science

Background:

• The exponential growth of digital data necessitates advanced data compression techniques for efficient storage and transmission.
• Current compression methods, including entropy, dictionary, predictive, and transform-based approaches, face limitations with increasingly large datasets.
• Genomic data, characterized by its unique structure and massive volume, presents specific challenges for effective compression.

Purpose of the Study:

• To introduce a novel encoding algorithm for data compression, inspired by the characteristics of DNA data.
• To address the limitations of existing compression methods by proposing a new approach for handling large-scale data.
• To demonstrate the efficacy of the proposed algorithm on genomic datasets and explore its applicability to other data types.

Main Methods:

• A divide-and-conquer strategy is employed, involving genome-wide scanning.
• Subsequences are classified based on content similarity and grouped into bins.
• Data within each bin is compressed independently, differentiating it from conventional methods.
• The algorithm's performance is validated using a benchmark dataset comprising seventeen diverse genomes.

Main Results:

• The novel algorithm achieved considerable improvements in genome compression ratios.
• Significant data size reduction was observed, preserving several megabytes compared to state-of-the-art tools.
• The proof-of-concept demonstrated the algorithm's effectiveness on genomes ranging from kilobytes to gigabytes.

Conclusions:

• The proposed encoding algorithm offers a substantial advancement in data compression, particularly for genomic data.
• Its unique binning approach provides superior compression efficiency over existing methods.
• The algorithm's versatility suggests potential applications in compressing diverse data types like text, images, audio, and video.