FPGA-Based Short Read Mapping for Genomic Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current software tools for short read mapping in genetic sequencing are bottlenecked by speed and require significant runtime, struggling to handle the volume of data and sensitivity needed for accurate genomic analysis, especially when dealing with genetic variations.
Innovation Solution
The use of field programmable gate arrays (FPGAs) for parallel and independent processing, which accelerates short read mapping by selecting seeds, determining candidate alignment locations, and calculating matching likelihoods, with techniques like Smith-Waterman string matching and partitioning of the reference genome index across multiple memories.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If software tools are used for short read mapping, then accuracy and sensitivity can be maintained, but runtime becomes excessively long and processing speed is bottlenecked
Solution Approach 1:
The reference genome index is partitioned across multiple memories, and the mapping process is divided into discrete stages (seed selection, candidate location determination, likelihood calculation) that can be executed in parallel. This segmentation enables the system to process multiple short reads simultaneously while maintaining mapping accuracy through systematic evaluation at each stage.
Solution Approach 2:
The patent replaces traditional software-based mapping mechanisms with hardware implementation using FPGAs. This substitution of mechanical/computational system enables parallel processing of multiple short reads against the reference genome, dramatically reducing runtime while preserving the algorithmic accuracy of methods like Smith-Waterman through dedicated hardware circuits.
2Productivity
If traditional processing methods are used, then system complexity remains manageable, but productivity and throughput are insufficient for large-scale genomic data
Solution Approach 1:
The patent transitions from sequential software processing to parallel hardware processing by utilizing the spatial dimension of FPGAs. Multiple mapping operations that would traditionally execute sequentially in software are performed simultaneously in hardware circuits, increasing throughput while the modular FPGA architecture keeps system complexity manageable through standardized building blocks.
Solution Approach 2:
The FPGA-based system performs mapping operations autonomously without requiring continuous software intervention. The hardware circuits self-manage the parallel processing of multiple short reads, automatically coordinating seed selection, candidate location identification, and likelihood calculation across the partitioned reference genome index stored in multiple memories.
3Measurement precision
If sensitivity is increased to detect genetic variations, then mapping accuracy improves, but processing time increases significantly
Solution Approach 1:
The reference genome index is pre-processed and partitioned across multiple memories before the actual mapping operation. This preliminary organization of reference data enables the FPGA system to quickly retrieve and compare candidate locations during processing, maintaining high sensitivity for detecting genetic variations while avoiding the time penalty of searching unorganized reference sequences during the mapping phase.
Data Source
AI summary
A scalable FPGA-based solution to the short read mapping problem in DNA sequencing is disclosed which greatly accelerates the task of aligning short length reads to a known reference genome. A representative system comprises one or more memory circuits storing a plurality of short reads and a reference genome sequence; and one or more field programmable gate arrays configured to select a short read; to extract a plurality of seeds from the short read, each seed comprising a genetic subsequence of the short read; for each seed, to determine at least one candidate alignment location (CAL) in the reference genome sequence to form a plurality of CALs; for each CAL, to determine a likelihood of the short read matching the reference genome sequence in the vicinity of the CAL; and to select one or more CALs having the currently greater likelihood of the short read matching the reference genome sequence.