BioVeL: a virtual laboratory for data analysis and modelling in biodiversity science and ecology

Abstract

Background: Making forecasts about biodiversity and giving support to policy relies increasingly on large collections of data held electronically, and on substantial computational capability and capacity to analyse, model, simulate and predict using such data. However, the physically distributed nature of data resources and of expertise in advanced analytical tools creates many challenges for the modern scientist. Across the wider biological sciences, presenting such capabilities on the Internet (as "Web services") and using scientific workflow systems to compose them for particular tasks is a practical way to carry out robust "in silico" science. However, use of this approach in biodiversity science and ecology has thus far been quite limited.

Results: BioVeL is a virtual laboratory for data analysis and modelling in biodiversity science and ecology, freely accessible via the Internet. BioVeL includes functions for accessing and analysing data through curated Web services; for performing complex in silico analysis through exposure of R programs, workflows, and batch processing functions; for on-line collaboration through sharing of workflows and workflow runs; for experiment documentation through reproducibility and repeatability; and for computational support via seamless connections to supporting computing infrastructures. We developed and improved more than 60 Web services with significant potential in many different kinds of data analysis and modelling tasks. We composed reusable workflows using these Web services, also incorporating R programs. Deploying these tools into an easy-to-use and accessible 'virtual laboratory', free via the Internet, we applied the workflows in several diverse case studies. We opened the virtual laboratory for public use and through a programme of external engagement we actively encouraged scientists and third party application and tool developers to try out the services and contribute to the activity.

Conclusions: Our work shows we can deliver an operational, scalable and flexible Internet-based virtual laboratory to meet new demands for data processing and analysis in biodiversity science and ecology. In particular, we have successfully integrated existing and popular tools and practices from different scientific disciplines to be used in biodiversity and ecological research.

Keywords: Analysis; Automation; Biodiversity science; Biodiversity virtual e-laboratory; Computing software; Data processing; Ecology; Informatics; Virtual laboratory; Workflows.

Fig. 1

Biodiversity virtual laboratory (BioVeL) is a software environment that assists scientists in collecting, organising, and sharing data processing and analysis tasks in biodiversity and ecological research. The main components of the platform are: A the Biodiversity Catalogue (a library with well-annotated data and analysis services); B the environment, such as RStudio for creating R programs; C the workbench for assembling data access and analysis pipelines; D the myExperiment workflow library that stores existing workflows; E the BioVeL Portal that allows researchers and collaborators to execute and share workflows; and F the documentation wiki. Infrastructure is indicated in bold, while processes related to research activities are indicated in italics. Components A–F are referred to from the text, where they are described in detail. See also ‘how-to’ guidelines in the Additional information