K-mer Based Biological Sequence Variant Characterization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for deriving typing information from next-generation sequencing (NGS) data, such as MLVA and MLST, are inefficient and lack reliable methods for determining repeat motif numbers and amplicon lengths, especially with the advent of high-throughput sequencing which generates large amounts of short, overlapping sequence reads.
Innovation Solution
The use of k-mer techniques to extract genomic features like primer delimited tandem repeat sequences and exact alleles from NGS data, involving error correction and the construction of de Bruijn graphs to infer repeat copy numbers and allele presence, enabling fast and accurate characterization of biological sequences.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional MLVA and MLST methods are used to derive typing information from NGS data, then genotyping can be performed, but the methods are inefficient and lack reliable methods for determining repeat motif numbers and amplicon lengths
Solution Approach 1:
The patent segments the NGS data analysis process into distinct computational steps: extracting k-mers from sequencing reads, counting k-mer frequencies, identifying repeat motifs through frequency patterns, and determining amplicon lengths. This segmentation enables reliable determination of repeat motif numbers by analyzing k-mer frequency distributions specific to tandem repeat regions, while maintaining computational efficiency through algorithmic optimization of each segment.
Solution Approach 2:
The patent performs preliminary actions by pre-processing NGS data to extract and count k-mers before actual genotyping analysis. K-mer frequency tables are constructed in advance, and repeat motifs are pre-identified through frequency thresholding. This preliminary processing enables rapid and reliable determination of repeat motif numbers and amplicon lengths during the actual typing process, significantly improving overall productivity.
2Productivity
If k-mer techniques are used to extract genomic features from NGS data, then the speed of MLVA and MLST analysis is improved, but complex computational processing is required
Solution Approach 1:
The patent replaces traditional mechanical sequencing and analysis methods with computational k-mer techniques. Instead of physical manipulation and manual analysis of genomic features, the system uses algorithms to extract k-mers, count frequencies, and identify repeat motifs computationally. This substitution dramatically increases analysis speed while the modular computational architecture manages complexity through systematic processing steps.
Solution Approach 2:
The patent changes the parameter of analysis from examining entire sequence contexts to analyzing fixed-length k-mer substrings. By transforming the data representation into k-mer frequency space and applying frequency thresholding parameters, the system achieves rapid identification of repeat motifs and alleles. This parameter transformation simplifies the computational problem by reducing sequence complexity to countable discrete units.
3Measurement precision
If error correction is applied in k-mer based analysis, then the accuracy of sequence characterization is improved, but additional processing steps are required
Solution Approach 1:
The patent implements self-service error correction by using the collective information from all sequencing reads to correct errors in individual reads. K-mers appearing at high frequencies are assumed to represent true genomic sequences, while low-frequency k-mers are identified as errors and corrected or discarded. This self-correcting mechanism improves sequence characterization accuracy using the inherent redundancy in NGS data without requiring external reference sequences or manual intervention.
Solution Approach 2:
The patent applies feedback by using k-mer frequency information to guide error correction decisions. The frequency distribution of k-mers provides feedback about which sequences are likely correct (high frequency) and which contain errors (low frequency). This feedback loop enables accurate error correction while minimizing processing time by automatically resolving ambiguities based on frequency thresholds without requiring iterative refinement.
Data Source
AI summary
Short fixed length sub-sequences, defined as reference sub-sequences, are extracted from a collection of reference sequences, and an index is constructed showing which short fixed length reference sub-sequence occurs in which reference sequences. Short fixed length sub-sequences, the same length as the reference sub-sequences and defined as source sub-sequences, are extracted from a collection of source sequences derived from a sample for which the signature is to be determined, and the short fixed length source sub-sequences are compiled to determine the frequency of each within the collection. The presence or absence of source sub-sequences in combination with the index is used to infer the presence or absence of reference sequences from the reference collection.