Dynamic Reference Graphs for Accurate Sequence Read Alignment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional mapping tools for biological sequence alignment, such as MAQ, BFAST, and BWA, often fail to accurately align sequence reads to a single static reference genome, leading to incorrect identification of genomic features and are computationally intensive.
Innovation Solution
A method and system that utilize a dynamic reference genome approach, incorporating a main path index and an alternate path index to align sequence reads, allowing for the detection of variants and updating the reference genome dynamically during alignment, thereby improving accuracy and efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional mapping tools align reads against a single static reference genome, then the alignment process is computationally intensive, but the accuracy of identifying genomic features deteriorates
Solution Approach 1:
The patent segments the reference genome into multiple alternative references representing different haplotypes or genomic variants. Instead of using a single static reference, the system creates a segmented reference structure that can represent multiple possible genomic configurations, allowing reads to be aligned against the most appropriate reference segment for their origin.
Solution Approach 2:
The patent implements a dynamic reference selection mechanism where the reference genome is not fixed but can be dynamically chosen or adjusted based on the characteristics of the reads being aligned. This dynamic approach allows the system to adapt the reference to match the sample being analyzed, improving accuracy without requiring exhaustive computational comparison against all possible references.
2Measurement precision
If conventional mapping tools use a single static reference genome, then the computational resources required are high, but the alignment accuracy deteriorates
Solution Approach 1:
The patent performs preliminary actions by pre-processing and organizing genomic variant information into a structured alternative reference format before alignment. This preliminary structuring of alternative references allows the alignment algorithm to efficiently query and compare against relevant references without having to process all possible genomic variations during the actual alignment step.
Solution Approach 2:
The patent changes the parameter of reference genome representation from a single static sequence to multiple alternative sequences with varying characteristics. By parameterizing the reference to include multiple haplotypes or variant configurations, the system can adjust which reference parameters are used based on the sample being analyzed, improving measurement precision while managing computational resources through selective parameter usage.
3Measurement precision
If high-depth sequencing is used to improve alignment accuracy, then the sequencing cost increases, but the need arises from conventional NGS inaccuracy
Solution Approach 1:
The patent introduces an intermediary layer between the raw sequencing reads and the final genomic analysis. This intermediary alternative reference structure acts as a mediator that captures and represents genomic diversity, allowing reads to be accurately mapped even at lower sequencing depths. The intermediary reference structure compensates for sequencing errors or ambiguities by providing multiple plausible alignment targets.
Solution Approach 2:
The patent creates copies of the reference genome in the form of alternative references representing different haplotypes or variant configurations. These copied and modified reference sequences allow the system to simulate multiple possible genomic realities without requiring multiple actual sequencing runs, effectively reducing the need for high-depth sequencing while maintaining accuracy.
Data Source
AI summary
A method for matching character strings to a reference character string, is disclosed. One or more processors receive a plurality of character strings. The one or more processors match each of the plurality of character strings to a main reference character string and registers a match to positions on the main reference character string that satisfy a pre-set match criteria. The one or more processors match each of the plurality of character strings to an alternate reference character string and registers a match to positions on the alternate reference character string that satisfy the pre-set match criteria. The alternate reference character string is derived from the main character string. The one or more processors identifies a match for each of the plurality of character strings that match to either a position on the main reference character string or the alternate reference character string.