Accelerating the Understanding of Life's Code Through Better Algorithms and Hardware Design

TL;DR

This paper introduces four novel pre-alignment filtering algorithms that leverage FPGA acceleration and CPU implementation to drastically reduce the computational time of genomic sequence alignment, significantly improving efficiency in genomic analysis.

Contribution

The paper presents four new algorithms for pre-alignment filtering that are highly accurate and efficiently accelerated using FPGA hardware, with a practical CPU implementation for broad applicability.

Findings

Hardware filters achieve 1000x speedup over CPU implementations.

Integration reduces read aligner execution time by up to 21.5x.

CPU implementation of SneakySnake reduces alignment time by up to 57.9x.

Abstract

Calculating the similarities between a pair of genomic sequences is one of the most fundamental computational steps in genomic analysis. This step -- called sequence alignment -- is the computational bottleneck because: (1) it is implemented using quadratic-time dynamic programming algorithms and (2) the majority of candidate locations in the reference genome do not align with a given read due to high dissimilarity. Calculating the alignment of such incorrect candidate locations consumes an overwhelming majority of a modern read mapper's execution time. In this thesis, we introduce four new algorithms (GateKeeper, Shouji, MAGNET, and SneakySnake) that function as a pre-alignment step and aim to filter out most incorrect candidate locations. The first key idea of our pre-alignment filters is to provide high filtering accuracy by correctly detecting all similar segments shared between two sequences. The second key idea is to exploit the massively parallel architecture of modern FPGAs for accelerating our filtering algorithms. We also develop an efficient CPU implementation of the SneakySnake algorithm for commodity desktops and servers. We evaluate the benefits and downsides of our pre-alignment filtering approach in detail using 12 real datasets. In our evaluation, we demonstrate that our hardware pre-alignment filters show two to three orders of magnitude speedup over their equivalent CPU implementations. We also demonstrate that integrating our hardware pre-alignment filters with the state-of-the-art read aligners reduces the aligner's execution time by up to 21.5x. Finally, we show that efficient CPU implementation of pre-alignment filtering still provides significant benefits. We show that SneakySnake on average reduces the execution time of the best performing CPU-based read aligners Edlib and Parasail, by up to 43x and 57.9x, respectively.