Predicting FQ Flashcards
What is the difference between bootstrapping and Monte Carlo simulation?
In bootstrapping, you randomly sample from your ORIGINAL data whereas in Monte Carlo you GENERATE random data from a known/assumed distribution
When can a reaction occur?
When the free energy change is negative
When is a reaction exothermic/endothermic?
When the change in enthalpy < 0, the reaction is exothermic
When the change in enthalpy > 0, the reaction is endothermic
What is the difference between a kinetically and a thermodynamically controlled reaction?
Kinetically controlled reactions depend on the availability of one of the reactants whereas thermodynamically controlled reactions are determined by the equilibrium position
What is the decimal reduction time?
the decimal reduction time D is the reciprocal of a rate constant
- it describes the time it takes to reduce the number of m.o. by a factor 10
What are TTIs?
Time-temperature Indicators: show a certain response as a function of time and temperature to which food is exposed
What is the working mechanism of TTIs?
such devices consist of enzymes that react with a substrate which leads to, for e.g., a color change
How do Bayesian statistics differ from classical statistics?
classical statistics = use of data
Bayesian statistics = use of data + prior knowledge
When can you expect a 0 order reaction to occur?
At the beginning of the reaction, too little breakdown of the compound of interest means the reaction rate equals the [c] pof the compound
What is the difference between an empirical and a mechanistic model?
Empirical = models that include fitted data
Mechanistic = determine the underlying physical/chemical mechanism
What equation describes the rate of an enzyme catalyzed reaction? What order does it follow?
Michaelis-Menten
first order (at low c) increases linearly and then 0 order (at high c) when it reached V max
Why is linearization of data affecting the quality of the data?
Because of linearization, the parameter values and their errors are also transformed, leading to
overestimation of the parameter values, and not a good fit
What is the equation for a batch reactor?
dc out /dt = r
there is no inflow nor outflow and the V is the same in the batch
What is the equation for CSTR?
accumulation is 0
0 = flow*(cin - cout)+rV
In a CSTR, what are the values of the concentrations in the formula?
enzyme activities related to the concentration in the vessel
if enzyme activity is reduced by 95%, the concentration in the vessel is 0.05
Which inactivation models are best for vegetative cells? for spore inactivation?
vegetative cells: Weibull model
spore inactivation: Shull or double Weibull
WHy is logarithmic transformation not good? In which case is it good?
Transformation of data that is normally distributed and homoscedastic disturbs the error structure of the data and the subsequent regression leads to bias.
For micro-organism data however, logarithmic transformation helps transform the data into a normal distribution (counting would give bias + lot of counting)
Describe main parameters in Weibull model?
- Contains 2 parameters: α (rate constant) and β (describing the shape of the curve)
- deals with vegetative cells
- In most cases, β > 1, suggesting that increasing damage occurs during heating
What is the Linewaver Burk plot?
transformation method: plots reciprocal value of the REACTION RATE vs the reciprocal value of [c]
How can you derive D (decimal reduction time) for the Weibull model?
alpha parameter of the Weibull model is D when beta=1
What is the Z-value?
the temperature increase needed to make D 10x smaller.
Can be obtained by plotting log (D) vs temp <=> only possible when the data are available at different temperatures
Pros/cons of Gompertz model in shelf-life predictions?
Pros:
- captures key phases (lag, exponential and stationary)
- simple parameters (lag time, growth rate, max population)
Cons:
- not mechanistic
- limited flexibility: does not perform well if conditions change
What is COP?
Cut-off point: when it is difficult to model the shelf-life of a product, consider a critical descriptor/factor determining shelf life
What’s the difference between deterministic and stochastic models?
deterministic = gives outcome as numerical value (same input=same output)
stochastic= gives an output with the associated TOTAL UNCERTAINTY (same input may result in different outputs)
