Fast short exact repeats finding on GPU

Repeat regions in DNA play very important roles in many vital biological functions. Repeats finding is always deemed as one of the most fundamental problems in genome sequencing and analysis, and exact repeats finding is the first step for many other repeats finding problems. This paper depicts the design and implementation issues of a fast data parallel short exact repeats finding algorithm on a commodity Graphic Processing Unit (GPU) using the Compute Unified Device Architecture (CUDA) platform, both from NVIDIA Corporation. The scalable parallel algorithm runs much faster for short DNA repeats without the loss of simplicity and generality, which can also be easily adapted for similar problems in proteome sequence analysis and other real-world text processing tasks.