Guest Editors’ Introduction: Hardware Acceleration in Computational Biology

h IT IS OUR pleasure to introduce this special issue on Hardware Acceleration in Computational Biology. The role of computing in molecular biology research has never been more defining. The synergy has reached a critical juncture where the rate of data generation is rapidly outpacing the rate at which it is processed. Most of the data processing for biocomputing applications currently is done in software, which takes a very long time. For example, aligning even hundreds of sequences using state-ofthe-art progressive alignment tools requires several hours on modern workstations. With sequencing technologies becoming increasingly high-throughput and increasingly commonplace at biologists’ research labs, large-scale sequence analysis, often involving millions to even tens of millions of sequences, has become one of the primary bottlenecks in the path to scientific discovery. The biocomputing domain also hosts a number of compute-intensive applications wherein the underlying problems are proven to be computationally intractable (e.g., phylogeny reconstruction). These aspects collectively position biocomputing as a domain that has the potential to immensely benefit through the incorporation of the latest advancements from the computing community. Several hardware accelerators to speed up data processing have been proposed recently. Among these, FPGA-based reconfigurable hardware platforms, Graphics Processing Unit (GPU), Cell Broadband Engine (CBE) and multicore processors are notable. Each of these has advantages and limitations. The principal advantages of using FPGA-, GPUor CBE-based systems are fast prototyping and ease of implementation. These systems primarily rely on software and use an existing hardware platform to map algorithms. On the other hand, the massive scale of fine-grain parallelism inherent in several biocomputing applications can be exploited efficiently in a multicore platform by integrating a large number of processing elements on a single chip. However, rapid prototyping with multicore processors is still not mature. The challenge of designing efficient hardware accelerators for biocomputing is actively being pursued by a number of researchers worldwide, and from a variety of different perspectives. Successful solutions will likely adopt and encompass elements from all or at least several levels of abstraction. This special issue highlights recent investigations regarding various hardware accelerators for biocomputing. The selected papers represent a design and benchmarking of wide range of accelerators, starting from field programmable gate arrays (FPGAs), to graphic processing units (GPUs), and multicore processors. ‘‘Hardware Accelerators in Computational Biology: Application, Potential, and Challenges,’’ by Majumder et al., presents a tutorial on the use of hardware platforms, such as FPGA, GPU, the Cell Broadband Engine (CBE), and custom multicore processors as accelerators. It also presents a