Genome Sequence Compression Through Mapping-Aware Read Encoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing genome sequencing data compression methods suffer from low compression ratios, slow processing speeds, information loss, and disruption of read order, which complicates downstream analysis and conformity checks.
Innovation Solution
A reference-based compression method that encodes genome sequence data by distinguishing perfectly and imperfectly mapped reads, using distinct encoding processes for each, and preserving the initial read order to ensure lossless compression and efficient decompression.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If reference-based compression methods are used, then compression ratio is improved, but processing speed deteriorates
Solution Approach 1:
The patent segments the compression process into three distinct encoding paths based on mapping quality: perfectly mapped reads (first encoding process), imperfectly mapped reads with few mismatches (third encoding process), and imperfectly mapped reads with many mismatches (second encoding process). This segmentation allows each segment to use an optimized encoding strategy, balancing compression ratio and processing speed for different read types.
Solution Approach 2:
The patent applies different encoding processes (local quality) to different subsets of reads based on their mapping characteristics. Perfectly mapped reads receive the most aggressive compression, while imperfectly mapped reads receive progressively less aggressive compression based on their mismatch count, optimizing the overall compression-performance tradeoff.
2Quantity of substance
If sequence alignment and classification methods are used, then compression ratio is improved, but information loss occurs
Solution Approach 1:
The patent performs preliminary classification of reads into three categories (perfectly mapped, imperfectly mapped with few mismatches, imperfectly mapped with many mismatches) before encoding, but preserves the original read order metadata. This preliminary action enables optimized compression while maintaining the ability to reconstruct the original sequence ordering.
Solution Approach 2:
The patent creates a compressed representation (copy) of the read data that encodes mapping quality information and sequence data in an optimized format, while maintaining the ability to reconstruct the original data exactly, thus preserving information while reducing storage requirements.
3Quantity of substance
If reads are reordered according to classes, then compression performance is improved, but downstream analysis reliability deteriorates
Solution Approach 1:
The patent extracts the read ordering information as separate metadata from the compressed sequence data itself. This allows the compression process to optimize encoding based on mapping quality classes while preserving the original read order information independently, ensuring downstream analysis reliability.
4Ease of operation
If general purpose text compression is used, then ease of operation is improved, but compression ratio deteriorates
Solution Approach 1:
The patent changes the encoding parameters based on the mapping quality of each read. Instead of using a single fixed compression scheme, the system dynamically selects from three different encoding processes based on mismatch count, optimizing compression ratio for each read type while maintaining operational simplicity through automated classification.
Data Source
Figure 1
Figure 1A
Figure 2A~2B
AI summary
The invention relates to a reference-based method for the compression of genome sequence data produced by a sequencing machine. The sequences of nucleotides or bases, that have been previously aligned to a reference sequence, are determined to be perfectly mapped, imperfectly mapped or unmapped with the reference sequence; and then coded according to said determination. The determining step comprises comparing, for each imperfectly mapped sequence, the number of mismatches between said sequence and the reference sequence with a reference threshold value, and encoding the imperfectly mapped sequences according to distinct encoding processes, depending on the result of said comparison method for the compression of genome sequence data produced by a sequencing machine.