Tackling the Challenges of FASTQ Referential Compression.

Area of Science:

• Bioinformatics
• Genomic Data Compression
• Computational Biology

Background:

• Exponential growth in genomic data strains bioinformatics storage.
• Referential compression offers higher potential than non-referential methods but remains underdeveloped.
• Efficient encoding of differences between input and reference data is key.

Purpose of the Study:

• Introduce a novel referential compression approach for FASTQ files.
• Develop an efficient encoding model for genomic data differences.
• Present the UdeACompress algorithm and its performance.

Main Methods:

• Developed a referential compressor combining local alignments with binary encoding.
• Optimized the binary encoding for long sequencing reads.
• Evaluated UdeACompress performance against state-of-the-art programs.

Main Results:

• UdeACompress achieved superior compression ratios for long reads.
• Demonstrated competitive compression ratios for shorter reads.
• Showcased reasonable execution times and memory consumption.

Conclusions:

• The novel UdeACompress algorithm effectively addresses the need for efficient genomic data compression.
• The approach shows significant promise for handling the increasing volume of sequencing data.
• UdeACompress offers a competitive solution in terms of compression efficiency and resource utilization.