high performance computing on graphics processing units: hgpu.org

Since the discovery of Deoxyribonucleic Acid (DNA) significant technological advances were made, leading to very large amounts of data gathered for analysis. The tools for this analysis however have advanced at a slower pace and have become one of the limiting factors of new discoveries in this field of research. Recently, from the 3D game market, a new generation of hardware has emerged. This hardware known as Graphics Processing Units (GPUs) has the capability for general purpose computing. With this new hardware high performance computing has become possible on cheap and readily available hardware which creates new opportunities to study and improve current tools and algorithms used for the study of DNA. A study of sequence matching techniques using indexes and their adaptation to the DNA alignment problem is presented. The indexes used were suffix trees and suffix arrays. A heuristic DNA alignment algorithm which runs on the GPU was developed based on the indexes studied. Tests conducted showed that the work lead to the creation of an algorithm that is capable of competing with current day techniques in both performance and quality of results.