Allele Identification via Pseudoalignment Indexing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for identifying alleles in genomic samples, such as HLA typing, are inefficient due to reliance on full genome alignment or de-novo assembly, which are slow and specific, and conventional SNP typing lacks resolution and transferability across populations.
Innovation Solution
The method employs an associative assembly procedure using an index of reference allele types, where paired-end sequencing reads are grouped and nucleotide substrings are extracted and matched against a reference data set, providing a pseudoalignment and probabilistic estimation of allele presence without the need for full genome alignment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If full genome alignment or de-novo assembly is used for allele identification, then measurement precision is improved, but productivity deteriorates due to slow processing speed
Solution Approach 1:
The patent extracts only the necessary nucleotide substrings (k-mers) from paired-end sequencing reads and compares them directly against a pre-built index of reference allele substrings. This extraction approach avoids the computationally intensive full genome alignment while maintaining the ability to accurately identify alleles by focusing only on the relevant discriminatory substrings that differentiate between alleles.
Solution Approach 2:
The patent performs preliminary action by pre-building an index of reference allele nucleotide substrings before the actual allele identification process. This index contains all possible k-mer substrings from reference alleles and their corresponding allele identifiers, allowing for rapid lookup and comparison during sequencing analysis, thereby eliminating the need for slow de-novo assembly or full alignment during the actual processing phase.
2Ease of operation
If conventional SNP typing is used for HLA typing, then ease of operation is improved, but measurement precision deteriorates due to insufficient resolution
Solution Approach 1:
The patent segments the allele sequences into multiple overlapping nucleotide substrings of length k (k-mers). By dividing the reference allele sequences into these smaller segments and building an index of all possible k-mers, the system can identify alleles by matching multiple segmented substrings from the sequencing reads against the index, achieving high resolution allele discrimination while maintaining operational simplicity through automated computational processing.
3Measurement precision
If a large number of SNP tags are typed to cover the sheer number of HLA alleles, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent creates a universal index structure that can handle any HLA allele in the reference database. The k-mer based index serves multiple functions: it stores all possible substrings from all reference alleles, enables rapid comparison against sequencing reads, and automatically adapts to any allele composition in the reference database. This universal approach eliminates the need for allele-specific assays or complex multi-plex systems, reducing overall system complexity while maintaining comprehensive allele coverage.
Data Source
AI summary
The present invention relates to systems and methods for determining the identity of alleles from genomic sequencing data via pseudoalignments. Particularly, aspects of the present invention are directed to a computer implemented method that includes obtaining a paired-end fragment from a genomic sample, determining a first nucleotide substring from a first end of the paired-end fragment matches a nucleotide substring within an index of nucleotide substrings and alleles, determining a second nucleotide substring from a second end of the paired-end fragment matches another nucleotide substring within the index, determining an intersection between the nucleotide substring and the another nucleotide substring exists, when an allele that contains the nucleotide substring is the same allele that contains the another nucleotide substring; and determining a probability that the paired end fragment is an observation of the allele based on the existence of the intersection.