Efficient lossless compression of nanopore sequencing signals

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• Efficient data compression is vital for managing large volumes of raw nanopore sequencing data, which present significant storage and transmission challenges.
• Existing compression methods often face limitations, either by increasing computational costs or by employing lossy techniques that compromise data integrity.

Purpose of the Study:

• To introduce PDZ, a novel lossless compression algorithm designed to enhance the efficiency of handling nanopore sequencing data.
• To demonstrate PDZ's superior performance in terms of compression ratio and computational speed compared to existing state-of-the-art methods.

Main Methods:

• Development of PDZ, a lossless compression algorithm implemented in C++.
• Integration of PDZ as a new compression method within the POD5 format.
• Comparative experimental evaluation against established compression algorithms like VBZ and EX-ZD.

Main Results:

• PDZ achieves compression ratio improvements of 0.87% to 2.84% over VBZ, depending on the dataset.
• PDZ demonstrates significant speedups: 1.09× to 2.25× faster compression and 1.01× to 1.52× faster decompression than VBZ.
• Compared to EX-ZD, PDZ offers a speedup factor of approximately 1.39× to 1.83× for both compression and decompression.

Conclusions:

• PDZ represents a significant advancement in lossless data compression for nanopore sequencing data.
• The algorithm offers a compelling balance of high compression performance and computational efficiency, outperforming current standards.
• PDZ is available as open-source, facilitating its adoption and further development within the bioinformatics community.