GateKeeper: a new hardware architecture for accelerating pre-alignment in DNA short read mapping.

Area of Science:

• Genomics
• Bioinformatics
• Computer Engineering

Background:

• High throughput DNA sequencing (HTS) generates vast amounts of short DNA reads, creating computational challenges for genome analysis.
• Mapping these short reads to a reference genome involves computationally intensive alignment, where most candidate locations are incorrect.
• Efficiently filtering these incorrect locations before alignment is critical to reduce computational bottlenecks.

Purpose of the Study:

• To develop a fast and effective hardware-based filter for pre-alignment in high throughput DNA sequencing.
• To accelerate the read mapping process by eliminating computationally expensive alignment of dissimilar sequences.

Main Methods:

• Proposed GateKeeper, a hardware accelerator utilizing Field-Programmable Gate Arrays (FPGAs) for pre-alignment filtering.
• Implemented GateKeeper as a pre-alignment step to filter incorrect candidate locations before computationally intensive alignment algorithms.

Main Results:

• GateKeeper achieves high accuracy (average >96%) in filtering incorrect candidate locations.
• GateKeeper provides significant speedups, averaging 90-fold over Adjacency Filter and 130-fold over Shifted Hamming Distance (SHD).
• Integration of GateKeeper reduced the verification time of the mrFAST mapper by a factor of 10.

Conclusions:

• GateKeeper offers a hardware acceleration solution for the pre-alignment bottleneck in DNA sequencing.
• The FPGA-based design demonstrates substantial performance gains and maintains high accuracy, improving overall read mapping efficiency.