WFA-GPU: gap-affine pairwise read-alignment using GPUs

• DOI: 10.1093/bioinformatics/btad701
• Publication Date: 2023-11-16T14:44:12Z
• Authors: Quim Aguado-Puig; Max Doblas; Christos Matzoros; Antonio Espinosa; Juan Carlos Moure; Santiago Marco-Sola; Miquel Moreto
• Keywords: Graphics processing unit

Advances in genomics and sequencing technologies demand faster more scalable analysis methods that can process longer sequences with higher accuracy. However, classical pairwise alignment methods, based on dynamic programming (DP), impose impractical computational requirements to align long noisy like those produced by PacBio Nanopore technologies. The recently proposed wavefront (WFA) algorithm paves the way for efficient tools, improving time memory complexity over previous methods. high-performance computing (HPC) platforms require parallel algorithms tools exploit resources available modern accelerator-based architectures.This paper presents WFA-GPU, a GPU (graphics processing unit)-accelerated tool compute exact gap-affine alignments WFA algorithm. We present algorithmic adaptations performance optimizations allow exploiting massively capabilities of devices accelerate computations. In particular, we propose CPU-GPU co-design capable performing inter-sequence intra-sequence sequence alignment, combining succinct WFA-data representation an implementation. As result, demonstrate our implementation outperforms original multi-threaded up 4.3× 18.2× when using heuristic sequences. Compared other state-of-the-art libraries, WFA-GPU is 29× than implementations four orders magnitude CPU implementations. Furthermore, only solution correctly aligning reads commodity GPU.WFA-GPU code documentation are publicly at https://github.com/quim0/WFA-GPU.