G-SQZ: compact encoding of genomic sequence and quality data.

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• High-throughput sequencing technologies generate vast amounts of data.
• Efficient data management, including storage, access, and transfer, is a significant challenge.
• Existing methods may not optimally address the specific nature of sequencing read data.

Purpose of the Study:

• To introduce G-SQZ, a novel data representation scheme for sequencing reads.
• To demonstrate the compression efficiency and selective access capabilities of G-SQZ.
• To highlight the practical benefits of G-SQZ for managing large sequencing datasets.

Main Methods:

• Development of G-SQZ, a Huffman coding-based compression algorithm tailored for sequencing reads.
• Implementation of the G-SQZ encoding scheme into software.
• Evaluation of G-SQZ performance on benchmark datasets.

Main Results:

• G-SQZ achieved significant data compression, ranging from 65% to 81% on benchmark datasets.
• The scheme allows for selective data access without requiring full sequential decoding.
• The software implementation is available for academic and non-profit use.

Conclusions:

• G-SQZ offers an effective solution for compressing large volumes of sequencing data.
• The method reduces infrastructure and informatics costs associated with data management and analysis.
• G-SQZ provides practical benefits for the genomics and bioinformatics fields.