Optimal compressed representation of high throughput sequence data via light assembly.

Area of Science:

• Bioinformatics
• Computational Biology
• Genomic Data Compression

Background:

• Effective genomic data compression is crucial for managing large datasets.
• Current leading methods rely on reference genomes or de novo assembly, demanding substantial computational resources.
• Faster, reference-free methods often sacrifice compression performance.

Purpose of the Study:

• To develop a novel reference-free genomic data compression method.
• To achieve high compression rates comparable to reference-based methods without the computational overhead.
• To improve the speed and efficiency of genomic data compression.

Main Methods:

• Introduced a new compressed representation for genomic data based on light de novo assembly.
• Represented each read as a node in a compact trie.
• Developed efficient algorithms for building these tries and compressing reads.

Main Results:

• The proposed method achieves the shortest possible output among all methods using the trie representation.
• The method approximates a theoretical lower bound on compression rate for uniformly sampled genomic data.
• Significantly improved compression performance compared to existing reference-free alternatives, without compromising speed.

Conclusions:

• The novel reference-free compression method offers a significant advancement in genomic data handling.
• It provides a balance between high compression efficiency and computational speed.
• This approach is a promising alternative for large-scale genomic data management.