Survey of Biological High Performance Computing: Algorithms, Implementations and Outlook Research

Abstract

During recent years there has been an explosive growth of biological data coming from genome projects, proteomics, protein structure determination, and the rapid expansion in digitization of patient biological data. Powerful computational techniques are required to understand and analyze biological information encoded by DNA sequences, which are frequently compared and searched for matching or near-matching patterns. Comparison of DNA sequences and genes can be useful to investigate the common functionalities of the corresponding organisms and to get a better understanding of how specific genes or groups of genes are organized. This kind of similarity calculation is known as sequence alignment and its objective is to identify similarities between subsequences of strings. Gene sequence alignment is one such problem that serves as an initial step in many of the problems in bioinformatics. Solving computational biology problems can be accelerated by algorithmic improvements or with the help of high-performance computing architectures. Such architectures include superscalar uniprocessors, parallel systems and dedicated hardware implementations of algorithms. FPGAs have emerged as high-performance computing accelerators, capable of implementing massively parallelized versions of computationally intensive algorithms. Their reprogrammability allows different algorithm-specific computing architectures to be implemented using the same hardware resource. In this article we provide a state of the art review for this field of research. We identify specific algorithmic problems and how hardware architectures can be designed to solve them. We present systems recently reported, describe their main features, and provide a comparison between them. Finally, we offer some directions for future investigations