A model for cereulide production in rice vs time at a fixed temperature allows for the quantification of toxin levels at different temperature ranges. This information is crucial for the assessment of the potential health risks associated with consuming cereulide-contaminated rice. It provides a quantitative basis for determining safe thresholds and establishing regulatory standards to protect public health.

The model helps identify the optimum temperature range for cereulide production by B. cereus in rice. It is anticipated that cereulide production will increase up to an optimal temperature and then decrease due to the denaturation of the synthetic machinery. Determining this optimum temperature range is vital for understanding the conditions under which cereulide formation is most favorable and implementing appropriate control measures.

Dynamic Temperature Environments: In real-world scenarios, such as the cooling of food after cooking, temperature fluctuations occur. Developing a model that considers the effect of temperature changes over time is essential. This dynamic perspective allows for a more accurate prediction of cereulide production in rice during various stages of food preparation, storage, and handling.


For our purposes, we used the model proposed by Baranyi and Roberts (1994) in their research of the growth rates of the B.cereus. They used the dynamic approach to predict the metabolic activity of bacteria in food. The significance of the lag phase kinetics is highly relevant in the field of food microbiology, whereas its importance in a bioreactor is comparatively diminished.


Here, represents the specific cereulide production rate expressed in ln(ng/g)/h, denotes the cereulide lag time in hours, signifies the initial cereulide level expressed in ng/g, and indicates the maximum cereulide level achieved, also expressed in ng/g.


Our primary goal was to fit the experimental data using this model and extend its application to a broader temperature range. Since we didn't have our own experimental data, we calibrated the model using parameter values obtained from the research article authored by Da Silva et al. (2022). Their study was conducted in the context of a dairy mix.

Furthermore, we wanted to calculate the time needed to produce a toxic dose of Cereulide for an average individual at a given temperature. Using these calculations, we would be able to determine the temperature at which the storage of the leftover rice is comparatively safe for consumption after a while. No doubts, it’s safer to store food at 12 degrees than 26 degrees Celsius.


The model was built in MATLAB, and we restricted the possibilities for the temperature range to 12-30 degrees Celsius as we cannot be sure that this model will work with high precision on a wider range.

Below is the data taken from the Da Silva’s research group:

Cereulide concentration (ng/g) and log(ng/g) vs time at different temperatures.

Cereulide Levels for Selected Temperatures

From the provided graphs, we observe a general trend indicating that higher temperatures result in increased Cereulide production. This can be attributed to the higher metabolic activity of B. cereus at warmer temperatures compared to cooler conditions.

The developed model can be utilized to predict whether leftover food poses a potential health risk or is relatively safe to consume. The toxic concentration for an average individual is approximately 1.6 micrograms per gram of meal (Jääskeläinen et al., 2003), equivalent to 1600 nanograms per gram (1600 ng/g). According to the model, at temperatures of 12°C and 15°C, the chances of spoilage are relatively low even if the food remains untouched for 100 hours (approximately 4 days). However, it is important to note that a bowl of rice under these moderately cool conditions may still harbor other potential pathogens.

Starting from 18°C, there is a definite risk of food poisoning if the food is left for more than 3.5 days, and this risk gradually increases with temperature. At 26°C and 30°C, it takes slightly over 2 days to reach the toxic concentration of Cereulide.

It is worth mentioning that the production graphs follow an exponential trend, and bacteria have their own limitations. Therefore, there is always a timeframe within which the deadly concentration of Cereulide will not be reached.


    [1] J. Baranyi and T. A. Roberts, “A dynamic approach to predicting bacterial growth in food,” International Journal of Food Microbiology, vol. 23, no. 3-4, pp. 277-294, 1994. doi:10.1016/0168-1605(94)90157-0
    [2] N. Buss da Silva et al., “Predicting B. cereus growth and Cereulide production in Dairy Mix,” International Journal of Food Microbiology, vol. 364, p. 109519, 2022. doi:10.1016/j.ijfoodmicro.2021.109519
    [3] Jääskeläinen, E.L. et al. (2003) ‘In vitro assay for human toxicity of cereulide, the emetic mitochondrial toxin produced by food poisoning bacillus cereus’, Toxicology in Vitro, 17(5-6), pp. 737-744. doi:10.1016/s0887-2333(03)00096-1.