C03055: Development of representative sampling plans for mycotoxins in foods using distribution modelling.

Thursday 5 October 2006

The aim of this project is to produce information on appropriate sampling strategies for the analysis of mycotoxins in raw food commodities using a computer modelling approach.

Study Duration : September 2006 - December 2007

Contractor : ADAS Ltd

Background

Studies have demonstrated that mycotoxins are seldom uniformly distributed throughout most food and feed commodities. Consequently, sampling plans to test for mycotoxin levels are complex and variable depending on the circumstances and the commodities involved.

In order to ensure consumer protection, food must be sampled in a representative way and although there is existing EU legislation on sampling for official controls, this is primarily designed for bulk sampling at points of entry. Therefore, this project aims to develop a model simulating the distribution of mycotoxins in food commodities which shows how the affected portions will be mixed and redistributed at various points of the food chain.

Research Approach

Initially, a review of the scientific literature will be carried out to identify sources of data that can be used to create the model (including the primary data associated with previous studies, sampling Directives and protocols).

Using the information gathered, a computer program will be written to produce a pilot model of distribution and sampling for one mycotoxin and one commodity. The model will generate probability distributions for the concentrations of toxins throughout a grain store, using at least two variables to characterise the distribution of the contamination in the model, including 1) the 'patch' size and 2) the 'patch concentration'.

Once the pilot model has been established, the effects of processing on mycotoxin distribution will be incorporated. Simple transformation of the original parameters will show the effect that mixing or separation of the commodity can have on the distribution of the contamination, such operations can cause concentration or dilution of the mycotoxin in the system.

Finally, the mechanism to test sampling strategies will be added to the model, established sampling protocols will be assess using the model and the effectiveness of new sampling schemes (e.g. the effect of varying the number of samples, their position and incremental sample size) will be tested.

Results and findings

The main conclusions were:

For most sample sizes, a regular grid proved to be more consistent and accurate in the estimation of the mean concentration of DON, which suggests that regular sampling strategies should be preferred to random sampling when possible.
For both regular and random sampling strategies, the accuracy of the estimation of the mean concentration increased significantly up to sample sizes of between 40 and 60 incremental samples (depending on the simulation). The effect of sample size was small when it exceeded 60 points, which suggests that the maximum sample size required is of this order;
These sample sizes are consistent with current recommendations for bulk cereals (60 incremental samples for 10-20 tonnes and 100 for 20-50 tonnes);
Similar conclusions applied to OTA, although the difference between regular and random sampling was small and probably negligible for most sample sizes.

This study was the first application of geostatistical analysis to date on mycotoxin contamination of bulk agricultural commodities.

Graphic version of this page

C03055: Development of representative sampling plans for mycotoxins in foods using distribution modelling.

Background

Research Approach

Results and findings

Tell a Friend

Printer friendly

Contact us

Get alerts

Our Sites