Read Mapping With Known Alleles on a Genomic DAG
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for analyzing next-generation sequencing (NGS) reads are computationally inefficient and fail to utilize the full potential of available genetic information, often relying on heuristic algorithms that sacrifice precision and limit the use of identified nucleotides, leading to misleading inferences when mapping to a single reference genome.
Innovation Solution
Utilizing a directed acyclic graph (DAG) that represents multiple alternative sequences at known locations, excluding incongruent mappings, and employing mathematically robust alignment algorithms like Smith-Waterman to align reads, thereby improving computational efficiency and accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If heuristic algorithms are used to map NGS reads to a reference genome, then computational speed is improved, but mathematical precision and accuracy are sacrificed
Solution Approach 1:
The patent pre-processes the reference genome to identify known variant positions (SNPs, indels, structural variants) and creates a curated reference structure with alternative sequences at these positions. This preliminary action allows the mapping algorithm to skip unnecessary comparisons at known variant sites, maintaining high speed while improving accuracy by considering multiple plausible alignments rather than forcing a single reference-matched alignment
Solution Approach 2:
The patent implements a dynamic mapping approach where the reference structure adapts to the specific sample being analyzed. By integrating sample-specific information (such as SNP array data or preliminary variant calls) into the mapping process, the algorithm dynamically adjusts which alternative sequences to consider at each position, optimizing both speed and precision for each individual case
2Device complexity
If a single linear reference genome is used for mapping, then simplicity is maintained, but accuracy is reduced due to misleading inferences from population-specific variants
Solution Approach 1:
The patent segments the reference genome into multiple alternative sequences at positions where population-specific variants are known to occur. Instead of a single linear reference, the structure is divided into reference sequences and alternative sequences that represent different population-specific haplotypes. This segmentation allows the mapping algorithm to choose the most appropriate reference segment for each region, improving accuracy without requiring the user to understand complex population genetics
Solution Approach 2:
The patent creates a universal reference structure that can serve multiple populations and ethnic groups simultaneously. By incorporating alternative sequences representing different population-specific variants into a single unified structure, the reference genome becomes multi-functional, accommodating diverse genetic backgrounds without requiring separate reference genomes for each population
3Reliability
If all possible sequence variations are considered during read mapping, then completeness is improved, but computational resources are exhausted
Solution Approach 1:
The patent applies local quality optimization by considering alternative sequences only at specific positions where population-specific variants are known to occur, rather than uniformly across the entire genome. This localized approach maintains mapping completeness at critical variant positions while avoiding the computational burden of considering all possible variations throughout the reference genome
Data Source
AI summary
The invention generally relates to genomic studies and specifically to improved methods for read mapping using identified nucleotides at known locations. The invention provides methods of using identified nucleotides at known places in a genome to guide the analysis of sequence reads from that genome by excluding potential mappings or assemblies that are not congruent with the identified nucleotides. Information about a plurality of SNPs in the subject's genome is used to identify candidate paths through a genomic directed acyclic graph (DAG). Sequence reads are mapped to the candidate paths.