Accelerating Edit-Distance Sequence Alignment on GPU Using the Wavefront Algorithm

Abstract

Sequence alignment remains a fundamental problem with practical applications ranging from pattern recognition to computational biology. Traditional algorithms based on dynamic programming are hard to parallelize, require significant amounts of memory, and fail to scale for large inputs. This work presents eWFA-GPU, a GPU (graphics processing unit)-accelerated tool to compute the exact editdistance sequence alignment based on the wavefront alignment algorithm (WFA). This approach exploits the similarities between the input sequences to accelerate the alignment process while requiring less memory than other algorithms. Our implementation takes full advantage of the massive parallel capabilities of modern GPUs to accelerate the alignment process. In addition, we propose a succinct representation of the alignment data that successfully reduces the overall amount of memory required, allowing the exploitation of the fast shared memory of a GPU. Our results show that our GPU implementation outperforms by 3-9× the baseline edit-distance WFA implementation running on a 20 core machine. As a result, eWFA-GPU is up to 265 times faster than state-of-the-art CPU implementation, and up to 56 times faster than state-of-the-art GPU implementations.

Gpu Acceleration of Levenshtein Distance Computation between Long Strings

“…The extend pyramid Ê is the pyramidal form of the extend table 𝐸 as defined in Eq. (8), which contains the length of the longest common prefixes of P and 𝑇 starting at positions 𝑖 and 𝑗 respectively.…”

“…When working with short strings, the workload is relatively small compared to the large number of processing elements (PEs) available on a GPU, resulting in low GPU occupancy. To address this, it is common practice to group a large number of short string pairs and process them in parallel, as described in [8]. However, when dealing with large strings, even with a pyramidal data structure, the amount of work D. Castells-Rufas Fig.…”

“…WFA was originally proposed to compute Smith-Waterman (SW) gap-affine global alignment [7] but it can be adapted to compute other distance or similarity scores based on computing a DP table. In [8], it was adapted to compute Levenshtein distance for limited strings lengths up to a thousand characters. The problem of using DP algorithms on very long strings has historically been hindered by the difficulty of managing the high memory demand and excessive computing time.…”

Performance Analysis of Parallel Implementations of Global Sequence Alignment Methods for DNA Sequencing

“…This particular property of DNA enables it to use in various fields [1]. DNA sequence matching is an important application used in forensics for identification purposes and in medicine and agriculture for exploring possible disease and abnormality diagnoses [2], [3]. A DNA sequence is composed of four letters: A, C, G, and T, as shown in Figure 1.…”

“…Some sequence alignment methods also use hardware-based or data-centric approaches that use special hardware or specific data structures such as FM-Index for faster processing [12], [13], [14]. However, most of the methods depend on algorithm-centric approaches such as edit distance, estimation algorithms, dynamic programming, etc., for fast processing [2], [15]. In the dynamic programming-based approach, a larger problem is divided into smaller independent sub-problems, and these sub-problems are solved individually [16].…”