FPGAs are increasing in computing power at a significant rate while the non-recurring engineering costs and time-to-market remain significant lower than those for application-specific integrated circuits (ASICs), encouraging FPGAs to be used in areas previous dominated by ASICs. In this study, we examine the appropriateness of FPGAs for high-performance, low-volume prodution parallel computing by mapping an existing ASIC-based massively parallel single-instruction, multiple data (SIMD) computer, the UCSC Kestrel, to a variety of FPGAs. The design has a raw peak performance of over 187 billion 8-bit operations per second (OPS), 48 times faster than the original ASIC-based Kestrel, using a Xilinx Virtex-6, and a cost efficiency of up to 81 MOPS/$ using a Xilinx Spartan-3. We also show that we can implement the entire original Kestrel (512 processing elements) as a system on a single programmable chip.

MOTIVATION: Data about biodiversity have been scattered in different formats in natural history collections, survey reports and the literature. A central challenge for the biodiversity informatics community is to provide the means to share and rapidly synthesize these data and the knowledge they provide us to build an easily accessible, unified global map of biodiversity. Such a map would provide raw and summary data and information on biodiversity and its change across the world at multiple scales. RESULTS: We discuss a series of steps required to create a unified global map of biodiversity. These steps include: building biodiversity repositories; creating scalable species distribution maps; creating flexible, user-programmable pipelines which enable biodiversity assessment; and integrating phylogenetic approaches into biodiversity assessment. We show two case studies that combine phyloinformatic and biodiversity informatic approaches to document large scale biodiversity patterns. The first case study uses data available from the Barcode of Life initiative in order to make species conservation assessment of North American birds taking into account evolutionary uniqueness. The second case study uses full genomes of influenza A available from Genbank to provide an auto-updating documentation of the evolution and geographic spread of these viruses. AVAILABILITY: Both the website for tracking evolution and spread of influenza A and the website for applying phyloinformatics analysis to Barcode of Life data are available as outcomes of case studies (http://biodiversity.colorado.edu).

BACKGROUND: Increasing the quantity and quality of data is a key goal of biodiversity informatics, leading to increased fitness for use in scientific research and beyond. This goal is impeded by a legacy of geographic locality descriptions associated with biodiversity records that are often heterogeneous and not in a map-ready format. The biodiversity informatics community has developed best practices and tools that provide the means to do retrospective georeferencing (e.g., the BioGeomancer toolkit), a process that converts heterogeneous descriptions into geographic coordinates and a measurement of spatial uncertainty. Even with these methods and tools, data publishers are faced with the immensely time-consuming task of vetting georeferenced localities. Furthermore, it is likely that overlap in georeferencing effort is occurring across data publishers. Solutions are needed that help publishers more effectively georeference their records, verify their quality, and eliminate the duplication of effort across publishers. RESULTS: We have developed a tool called BioGeoBIF, which incorporates the high throughput and standardized georeferencing methods of BioGeomancer into a beginning-to-end workflow. Custodians who publish their data to the Global Biodiversity Information Facility (GBIF) can use this system to improve the quantity and quality of their georeferences. BioGeoBIF harvests records directly from the publishers' access points, georeferences the records using the BioGeomancer web-service, and makes results available to data managers for inclusion at the source. Using a web-based, password-protected, group management system for each data publisher, we leave data ownership, management, and vetting responsibilities with the managers and collaborators of each data set. We also minimize the georeferencing task, by combining and storing unique textual localities from all registered data access points, and dynamically linking that information to the password protected record information for each publisher. CONCLUSION: We have developed one of the first examples of services that can help create higher quality data for publishers mediated through the Global Biodiversity Information Facility and its data portal. This service is one step towards solving many problems of data quality in the growing field of biodiversity informatics. We envision future improvements to our service that include faster results returns and inclusion of more georeferencing engines.