FastQ/FastA compression systems and methods

Inventors

Nazari, Foad • Patel, Sneh • Murray, Emma K. • Schena, Giana J.

Assignees

Rajant Health Inc

Abstract

Systems and methods to analyze and significantly compress FastQ and/or FastA datasets are disclosed. The methodology includes algorithms to compress sequences, quality scores and identifiers of read files. The method relies on reducing the dimension and redundancy in genomic data in a unique and optimal way and in the binary format. The methodology also includes the decoding protocols to decompress the compressed data with zero loss.

Core Innovation

The disclosed invention relates to data compression of genomic data by receiving a data file comprised of sequence bases, quality scores, and identifiers, where the sequence bases comprise regular bases adenine (A), cytosine (C), guanine (G), and thymine (T) and irregular bases. The method applies an optimization algorithm to the quality scores, sequence k-mers from the sequence bases, and the identifiers, and splits long reads of the sequence bases and the quality scores into smaller segments.

After read splitting and deletion of duplicated and semi-duplicated reads, the invention performs dimensionality reduction on the sequence bases. It stores a template of the identifier that is consistent across the data file, and it detects and stores the location and type of each irregular base.

The encoding step represents each regular base by one of four two-digit binary numbers and represents all irregular bases by one of four two-digit binary numbers, while maintaining the irregular-base location and type information. The encoded data file is then compressed, where compression is lossless and reference-free in at least some embodiments described in the document, and the described compression supports FastQ and FastA genomic data file types.

Claims Coverage

The partial content provides two independent claims: one method claim and one non-transitory computer storage medium claim. Each independent claim specifies a chain of inventive features totaling receiving and preprocessing genomic sequence, quality score, and identifier data, dimensionality reduction and identifier templating, irregular-base location and type tracking with binary encoding, and lossless compression.

Across both independent claims, the inventive coverage centers on jointly optimizing quality scores, sequence k-mers, and identifiers; preprocessing by splitting long reads and deleting duplicated and semi-duplicated reads; applying dimensionality reduction; using a consistent identifier template; detecting and storing irregular-base location and type; encoding regular and irregular bases into binary representations using four two-digit binary numbers each; and performing lossless compression.

Stated Advantages

Documented Applications