The database combines two major software packages. The first is ARS's Pathogen Modelling Programme software, a research and instructional tool for estimating the effects of multiple variables on growth, inactivation, or survival of food borne pathogens. This has been available for download since 2002.
The second is the United Kingdom's Food MicroModel, produced by IFR and the Foods Standards Agency, which is used to describe bacterial responses to food environments.
By combining their efforts through the establishment of ComBase, both the ARS and IFR hope to make a large body of information accessible to as many food researchers and scientists as possible. Scientists can use the database to enter data such as the temperature, acidity and available water, and then retrieve all records that match the search criteria.
The database already contains about 25,000 growth and survival data records. Microbiologists in academia, government and industry can continue to submit data to ComBase, thus eliminating unnecessary repetition of experiments among laboratories, improving models, and standardising data sources.
Indeed according to project coordinator Mark Tamplin, the ultimate success of ComBase relies on collaborators' willingness to deposit more data.
"Development of models depends on available data," he said. "We're appealing to professional journals to ask authors to submit all their raw data with their articles, much as they already do for papers about genomic sequences. This would help keep the database timely and offer users the most reliable information."
Predictive microbiology is a growing field that estimates behaviour of micro organisms in response to environmental conditions, including food production and processing operations from the farm to the table. Indeed, the ARS has been developing mathematical models of the behaviour of bacterial pathogens in food for the past 15 years.
In February 2002, the service's Eastern Regional Research Centre (ERRC) established the Centre of Excellence in Microbial Modelling and Informatics (CEMMI) to help generate interest in forming partnerships that advance use of predictive models of micro organisms in food.
CEMMI's objective is to link the expertise of its members to researchers in the food industry, government, and academia. According to Tamplin, the centre hopes to improve the way predictive models are developed and then applied to various food-processing situations, while ensuring that users properly interpret results.