Package {predmicror}

Title:	Fitting Predictive Microbiology Models
Version:	1.3.2
Date:	2026-06-05
Description:	Provides predictive microbiology model functions and convenience wrappers for fitting primary growth, microbial inactivation, dynamic, omnibus, and cardinal parameter models to experimental data using nonlinear least squares and related mixed-effects or time-varying workflows. Includes helper functions for extracting fitted values, calculating model diagnostics, and comparing fitted models. Implemented model families include those described by: Zwietering et al. (1990) <doi:10.1128/AEM.56.6.1875-1881.1990>, Baranyi and Roberts (1994) <doi:10.1016/0168-1605(94)90157-0>, Baranyi and Roberts (1995) <doi:10.1016/0168-1605(94)00121-L>, Buchanan et al. (1997) <doi:10.1006/fmic.1997.0125>, Richards (1959) <doi:10.1093/jxb/10.2.290>, Fang et al. (2012) <doi:10.1111/j.1750-3841.2012.02873.x>, Fang et al. (2013) <doi:10.1016/j.fm.2012.12.005>, Huang (2008) <doi:10.1111/j.1750-3841.2008.00785.x>, Huang (2009) <doi:10.1016/j.jfoodeng.2008.07.011>, Huang (2013) <doi:10.1016/j.foodcont.2012.11.019>, Geeraerd et al. (2005) <doi:10.1016/j.ijfoodmicro.2004.11.038>, van Boekel (2002) <doi:10.1016/S0168-1605(01)00742-5>, Peleg (1999) <doi:10.1016/S0963-9969(99)00081-2>, Mafart et al. (2002) <doi:10.1016/S0168-1605(01)00624-9>, Albert and Mafart (2005) <doi:10.1016/j.ijfoodmicro.2004.10.016>, Rosso et al. (1993) <doi:10.1006/jtbi.1993.1099>, Rosso et al. (1995) <doi:10.1128/AEM.61.2.610-616.1995>, and Rosso et al. (1996) <doi:10.4315/0362-028X-59.9.944>.
Depends:	R (≥ 3.4.0)
Imports:	graphics, gslnls, nlme, Rdpack, stats, utils
RdMacros:	Rdpack
License:	GPL (≥ 3)
LazyLoad:	yes
LazyData:	true
Encoding:	UTF-8
RoxygenNote:	7.3.3
NeedsCompilation:	no
Suggests:	knitr, rmarkdown, htmltools, commonmark, shinyAce, testthat (≥ 3.0.0), ggplot2, deSolve, FME, minpack.lm, reshape2, readxl, shiny
VignetteBuilder:	knitr
URL:	https://fsqanalytics.github.io/predmicror/, https://github.com/fsqanalytics/predmicror
BugReports:	https://github.com/fsqanalytics/predmicror/issues
Config/testthat/edition:	3
Packaged:	2026-06-13 22:58:17 UTC; vcadavez
Author:	Vasco Cadavez [aut, cre], Ursula Gonzales-Barron [aut]
Maintainer:	Vasco Cadavez <vcadavez@ipb.pt>
Repository:	CRAN
Date/Publication:	2026-06-19 16:30:02 UTC

predmicror: Fitting Predictive Microbiology Models

Description

Provides predictive microbiology model functions and convenience wrappers for fitting primary growth, microbial inactivation, dynamic, omnibus, and cardinal parameter models to experimental data using nonlinear least squares and related mixed-effects or time-varying workflows. Includes helper functions for extracting fitted values, calculating model diagnostics, and comparing fitted models. Implemented model families include those described by: Zwietering et al. (1990) doi:10.1128/AEM.56.6.1875-1881.1990, Baranyi and Roberts (1994) doi:10.1016/0168-1605(94)90157-0, Baranyi and Roberts (1995) doi:10.1016/0168-1605(94)00121-L, Buchanan et al. (1997) doi:10.1006/fmic.1997.0125, Richards (1959) doi:10.1093/jxb/10.2.290, Fang et al. (2012) doi:10.1111/j.1750-3841.2012.02873.x, Fang et al. (2013) doi:10.1016/j.fm.2012.12.005, Huang (2008) doi:10.1111/j.1750-3841.2008.00785.x, Huang (2009) doi:10.1016/j.jfoodeng.2008.07.011, Huang (2013) doi:10.1016/j.foodcont.2012.11.019, Geeraerd et al. (2005) doi:10.1016/j.ijfoodmicro.2004.11.038, van Boekel (2002) doi:10.1016/S0168-1605(01)00742-5, Peleg (1999) doi:10.1016/S0963-9969(99)00081-2, Mafart et al. (2002) doi:10.1016/S0168-1605(01)00624-9, Albert and Mafart (2005) doi:10.1016/j.ijfoodmicro.2004.10.016, Rosso et al. (1993) doi:10.1006/jtbi.1993.1099, Rosso et al. (1995) doi:10.1128/AEM.61.2.610-616.1995, and Rosso et al. (1996) doi:10.4315/0362-028X-59.9.944.

Author(s)

Maintainer: Vasco Cadavez vcadavez@ipb.pt (ORCID)

Authors:

• Ursula Gonzales-Barron ubarron@ipb.pt (ORCID)

See Also

Useful links:

• https://fsqanalytics.github.io/predmicror/

• https://github.com/fsqanalytics/predmicror

• Report bugs at https://github.com/fsqanalytics/predmicror/issues

Baranyi and Roberts full growth model

Description

BaranyiFM function to fit the Baranyi & Roberts full growth model to a complete microbial growth curve. Returns the model parameters estimated according to data collected in microbial growth experiments.

Usage

BaranyiFM(t, Y0, Ymax, MUmax, lag)

Arguments

Details

The model's inputs are:

t: time, assuming time zero as the beginning of the experiment.

Y(t): the natural logarithm of the microbial concentration (ln(N(t))) measured at time t.

Users should make sure that the microbial concentration input is entered in natural logarithm, Y(t) = ln(N(t)).

Value

A numeric vector with the fitted values

Author(s)

Vasco Cadavez vcadavez@ipb.pt and Ursula Gonzales-Barron ubarron@ipb.pt

References

Baranyi J, Roberts TA (1994). “A dynamic approach to predicting bacterial growth in food.” International Journal of Food Microbiology, 23, 277-294.

Examples

library(gslnls)
data(growthfull)
initial_values <- list(Y0 = -0.1, Ymax = 22, MUmax = 1.7, lag = 5)
fit <- gsl_nls(lnN ~ BaranyiFM(Time, Y0, Ymax, MUmax, lag),
  data = growthfull,
  start = initial_values
)
summary(fit)

Baranyi and Roberts reduced growth model

Description

BaranyiRM function to fit the Baranyi and Roberts growth model to a reduced microbial growth curve. Returns the model parameters estimated according to data collected in microbial growth experiments.

Usage

BaranyiRM(t, Y0, MUmax, lag)

Arguments

Details

Model's inputs are:

t: time, assuming time zero as the beginning of the experiment.

Y(t): the natural logarithm of the microbial concentration (ln(N(t))) measured at time t.

Users should make sure that the microbial concentration input is entered in natural logarithm, Y(t) = ln(N(t)).

Value

A numeric vector with the fitted values

Author(s)

Vasco Cadavez, vcadavez@ipb.pt and Ursula Gonzales-Barron, ubarron@ipb.pt

References

Baranyi J, Roberts TA (1995). “Mathematics of predictive microbiology.” International Journal of Food Microbiology, 26, 199-218.

Examples

## Example: Baranyi reduced model
library(gslnls)
data(growthred) # simulated data set.
initial_values <- list(Y0 = 0.1, MUmax = 1.7, lag = 5) # define the initial values
fit <- gsl_nls(lnN ~ BaranyiRM(Time, Y0, MUmax, lag),
  data = growthred,
  start = initial_values
)
summary(fit)

Buchanan reduced growth model

Description

BuchananRM function to fit the Buchanan reduced growth model to a reduced microbial growth curve. Returns the model parameters estimated according to data collected in microbial growth experiments.

Usage

BuchananRM(t, Y0, MUmax, lag)

Arguments

Details

Model's inputs are:

t: time, assuming time zero as the beginning of the experiment.

Y(t): the natural logarithm of the microbial concentration (ln(N(t))) measured at time t.

Users should make sure that the microbial concentration input is entered in natural logarithm, Y(t) = ln(N(t)).

Value

A numeric vector with the fitted values

Author(s)

Vasco Cadavez, vcadavez@ipb.pt and Ursula Gonzales-Barron, ubarron@ipb.pt

References

Buchanan RL, Whiting RC, Damert WC (1997). “When is simple good enough: a comparison of the Gompertz, Baranyi, and three-phase linear models for fitting bacterial growth curves.” Food Microbiology, 14(4), 313-326. ISSN 0740-0020, doi:10.1006/fmic.1997.0125.

Examples

## Example: Buchanan reduced model
library(gslnls)
data(growthred) # simulated data set.
initial_values <- list(Y0 = 0, MUmax = 1.7, lag = 5) # define the initial values
fit <- gsl_nls(lnN ~ BuchananRM(Time, Y0, MUmax, lag),
  data = growthred,
  start = initial_values
)
summary(fit)

Cardinal model for water activity

Description

CMAW function to fit the water activity cardinal model (Rosso et al., 1993). Returns the model parameters estimated according to data collected in microbial growth experiments.

Usage

CMAW(x, AWmin, MUopt, AWopt)

Arguments

Details

The model's inputs are:

x: Water activity

sqrtGR: the square root of the growth rate (h^{-1})

Users should make sure that the growth rate input is entered after a square root transformation, sqrGR = sqrt(GR).

Value

A numeric vector with the fitted values

Author(s)

Vasco Cadavez vcadavez@ipb.pt and Ursula Gonzales-Barron ubarron@ipb.pt

References

Rosso L, Lobry J, Charles (Bajard) S, Flandrois J (1995). “Convenient Model To Describe the Combined Effects of Temperature and pH on Microbial Growth.” Applied and environmental microbiology, 61, 610–6. doi:10.1128/AEM.61.2.610-616.1995.

Examples

library(gslnls)
data(aw)
initial_values <- list(AWmin = 0.89, MUopt = 1.0, AWopt = 0.98)
fit <- gsl_nls(sqrtGR ~ CMAW(aw, AWmin, MUopt, AWopt),
  data = aw,
  start = initial_values
)
summary(fit)

Cardinal model for growth inhibitors

Description

CMInh function to fit the growth inhibitors cardinal model (Rosso et al, 1993). Returns the model parameters estimated according to data collected in microbial growth experiments.

Usage

CMInh(x, MIC, MUopt, alpha)

Arguments

Details

The model's inputs are:

x: growth inhibitor concentration

sqrtGR: the square root of the growth rate (time^{-1})

Users should make sure that the growth rate input is entered after a square root transformation, $sqrGR = sqrt(GR)$.

Value

A numeric vector with the fitted values

Author(s)

Vasco Cadavez vcadavez@ipb.pt and Ursula Gonzales-Barron ubarron@ipb.pt

References

Rosso L, Lobry J, Charles (Bajard) S, Flandrois J (1995). “Convenient Model To Describe the Combined Effects of Temperature and pH on Microbial Growth.” Applied and environmental microbiology, 61, 610–6. doi:10.1128/AEM.61.2.610-616.1995.

Examples

library(gslnls)
data(inh)
initial_values <- list(MIC = 0.89, MUopt = 1.0, alpha = 1)
fit <- gsl_nls(sqrtGR ~ CMInh(Conce, MIC, MUopt, alpha),
  data = inh,
  start = initial_values
)
summary(fit)

Cardinal model for pH

Description

CMPH function to fit the pH cardinal model (Rosso et al, 1993). Returns the model parameters estimated according to data collected in microbial growth experiments.

Usage

CMPH(x, pHmax, pHmin, MUopt, pHopt)

Arguments

Details

The model's inputs are:

x: pH

sqrtGR: the square root of the growth rate (time^{-1})

Users should make sure that the growth rate input is entered after a square root transformation, sqrGR = sqrt(GR).

Value

A numeric vector with the fitted values

Author(s)

Vasco Cadavez vcadavez@ipb.pt and Ursula Gonzales-Barron ubarron@ipb.pt

References

Rosso L, Lobry J, Charles (Bajard) S, Flandrois J (1995). “Convenient Model To Describe the Combined Effects of Temperature and pH on Microbial Growth.” Applied and environmental microbiology, 61, 610–6. doi:10.1128/AEM.61.2.610-616.1995.

Examples

library(gslnls)
data(ph)
initial_values <- list(pHmax = 9, pHmin = 3, MUopt = 1.0, pHopt = 7)
fit <- gsl_nls(sqrtGR ~ CMPH(pH, pHmax, pHmin, MUopt, pHopt),
  data = ph,
  start = initial_values
)
summary(fit)

Cardinal model for temperature

Description

CMTI function to fit the temperature cardinal model (Rosso et al, 1993). Returns the model parameters estimated according to data collected in microbial growth experiments.

Usage

CMTI(x, Tmax, Tmin, MUopt, Topt)

Arguments

Details

The model's inputs are:

x: Temperature

sqrtGR: the square root of the growth rate (h^{-1})

Users should make sure that the growth rate input is entered after a square root transformation, sqrGR = sqrt(GR).

Value

A numeric vector with the fitted values

Author(s)

Vasco Cadavez vcadavez@ipb.pt and Ursula Gonzales-Barron ubarron@ipb.pt

References

Rosso L, Lobry J, Charles (Bajard) S, Flandrois J (1995). “Convenient Model To Describe the Combined Effects of Temperature and pH on Microbial Growth.” Applied and environmental microbiology, 61, 610–6. doi:10.1128/AEM.61.2.610-616.1995.

Examples

library(gslnls)
data(salmonella)
initial_values <- list(Tmax = 42, Tmin = 1, MUopt = 1.0, Topt = 37)
fit <- gsl_nls(sqrtGR ~ CMTI(Temp, Tmax, Tmin, MUopt, Topt),
  data = salmonella,
  start = initial_values
)
summary(fit)

plot(salmonella$Temp, salmonella$sqrtGR^2)
lines(salmonella$Temp, fitted(fit)^2, col = "green")
plot(salmonella$Temp, salmonella$sqrtGR)
lines(salmonella$Temp, fitted(fit), col = "red")

Fang no lag growth model

Description

FangNLM function to fit the Fang no lag growth model to an incomplete microbial growth curve. Returns the model parameters estimated according to data collected in microbial growth experiments.

Usage

FangNLM(t, Y0, Ymax, MUmax)

Arguments

Details

Model's inputs are:

t: time, assuming time zero as the beginning of the experiment.

Y(t): the natural logarithm of the microbial concentration (ln(N(t))) measured at time t.

Users should make sure that the microbial concentration input is entered in natural logarithm, Y(t) = ln(N(t)).

Value

A numeric vector with the fitted values

Author(s)

Vasco Cadavez, vcadavez@ipb.pt and Ursula Gonzales-Barron, ubarron@ipb.pt

References

Fang T, Gurtler JB, Huang L (2012). “Growth Kinetics and Model Comparison of Cronobacter sakazakii in Reconstituted Powdered Infant Formula.” Journal of Food Science, 77(9), E247-E255. doi:10.1111/j.1750-3841.2012.02873.x. Fang T, Liu Y, Huang L (2013). “Growth kinetics of Listeria monocytogenes and spoilage microorganisms in fresh-cut cantaloupe.” Food Microbiology, 34(1), 174-181. ISSN 0740-0020, doi:10.1016/j.fm.2012.12.005.

Examples

## Example: Fang no lag model
library(gslnls)
data(growthnolag) # simulated data set.
initial_values <- list(Y0 = 0, Ymax = 22, MUmax = 1.7) # define the initial values
fit <- gsl_nls(lnN ~ FangNLM(Time, Y0, Ymax, MUmax),
  data = growthnolag,
  start = initial_values
)
summary(fit)

Geeraerd inactivation model

Description

GeeraerdST inactivation model for microbial inactivation curve. Returns the model parameters estimated according to data collected in microbial inactivation experiments.

Usage

GeeraerdST(x, Y0, Yres, kmax, Sl)

Arguments

Details

The model's inputs are:

t: time, assuming time zero as the beginning of the experiment.

N(t): the bacterial concentration measured at time t.

Users should make sure that the bacterial concentration input is entered in natural logarithm, Y(t) = ln(N(t)).

Value

A numeric vector with the fitted values

Author(s)

Vasco Cadavez vcadavez@ipb.pt and Ursula Gonzales-Barron ubarron@ipb.pt

References

Geeraerd AH, Valdramidis VP, Van Impe JF (2005). “GInaFiT, a freeware tool to assess non-log-linear microbial survivor curves.” International Journal of Food Microbiology, 102(1), 95-105. ISSN 0168-1605, doi:10.1016/j.ijfoodmicro.2004.11.038.

Examples

library(gslnls)
data(mafart2005Li11)
mafart2005Li11$lnN <- log(10) * mafart2005Li11$logN
initial_values <- list(Y0 = 18, Yres = 2, kmax = 0.7, Sl = 4)
fit <- gsl_nls(lnN ~ GeeraerdST(Time, Y0, Yres, kmax, Sl),
  data = mafart2005Li11,
  start = initial_values
)
summary(fit)

plot(lnN ~ Time, data = mafart2005Li11)
lines(mafart2005Li11$Time, predict(fit), col = "blue")

Huang full growth model

Description

HuangFM function to fit the Huang full growth model to complete microbial growth curve. Returns the model parameters estimated according to data collected in microbial growth experiments.

Usage

HuangFM(t, Y0, Ymax, MUmax, lag)

Arguments

Details

Model's inputs are:

t: time, assuming time zero as the beginning of the experiment.

Y(t): the natural logarithm of the microbial concentration (ln(N(t))) measured at time t.

Users should make sure that the microbial concentration input is entered in natural logarithm, Y(t) = ln(X(t)).

Value

A numeric vector with the fitted values

Author(s)

Vasco Cadavez, vcadavez@ipb.pt and Ursula Gonzales-Barron, ubarron@ipb.pt

References

Huang L (2008). “Growth Kinetics of Listeria monocytogenes in Broth and Beef Frankfurters-Determination of Lag Phase Duration and Exponential Growth Rate under Isothermal Conditions.” Journal of Food Science, 73(5), E235-E242. doi:10.1111/j.1750-3841.2008.00785.x.

Examples

## Example: Huang full model
library(gslnls)
data(growthfull) # simulated data set.
initial_values <- list(Y0 = 0, Ymax = 22, MUmax = 1.7, lag = 5) # define the initial values
## Call the fitting function
fit <- gsl_nls(lnN ~ HuangFM(Time, Y0, Ymax, MUmax, lag),
  data = growthfull,
  start = initial_values
)
summary(fit)

confint(fit)

preds <- data.frame(predict(fit, interval = "prediction", level = 0.95))
plot(lnN ~ Time, data = growthfull, ylim = c(-1, 22))
lines(growthfull$Time, preds$fit, col = "blue")
lines(growthfull$Time, preds$upr, col = "red")
lines(growthfull$Time, preds$lwr, col = "red")

Huang no lag growth model

Description

HuangNLM function to fit the Huang no lag growth model to an incomplete microbial growth curve. Returns the model parameters estimated according to data collected in microbial growth experiments.

Usage

HuangNLM(t, Y0, Ymax, MUmax)

Arguments

Details

Model's inputs are:

t: time, assuming time zero as the beginning of the experiment.

Y(t): the natural logarithm of the microbial concentration (ln(N(t)) measured at time t.

Users should make sure that the microbial concentration input is entered in natural logarithm, Y(t) = ln(N(t)).

Value

A numeric vector with the fitted values

Author(s)

Vasco Cadavez, vcadavez@ipb.pt and Ursula Gonzales-Barron, ubarron@ipb.pt

References

Huang (2013).

Examples

## Example: Huang no lag model
library(gslnls)
data(growthnolag) # simulated data set.
initial_values <- list(Y0 = 0, Ymax = 22, MUmax = 1.7) # define the initial values
## Call the fitting function
fit <- gsl_nls(lnN ~ HuangNLM(Time, Y0, Ymax, MUmax),
  data = growthnolag,
  start = initial_values
)
summary(fit)

Huang reparameterized Gompertz survival model

Description

HuangRGS reparametrized Gompertz survival model for microbial inactivation. Returns the model parameters estimated according to data collected in microbial inactivation experiments.

Usage

HuangRGS(x, Y0, k, M)

Arguments

Details

The model's inputs are:

t: time, assuming time zero as the beginning of the experiment.

Y(t): the natural logarithm of the bacterial concentration (ln(X(t))) measured at time t.

Users should make sure that the bacterial concentration input is entered in natural logarithm, Y(t) = ln(X(t)).

Value

A numeric vector with the fitted values

Author(s)

Vasco Cadavez vcadavez@ipb.pt and Ursula Gonzales-Barron ubarron@ipb.pt

References

Huang L (2009). “Thermal inactivation of Listeria monocytogenes in ground beef under isothermal and dynamic temperature conditions.” Journal of Food Engineering, 90(3), 380-387. ISSN 0260-8774, doi:10.1016/j.jfoodeng.2008.07.011.

Examples

library(gslnls)
data(bixina)
initial_values <- list(Y0 = 5.6, k = 0.37, M = 6.8)
fit <- gsl_nls(lnN ~ HuangRGS(Time, Y0, k, M),
  data = bixina,
  start = initial_values
)
summary(fit)

plot(lnN ~ Time, data = bixina)
lines(bixina$Time, predict(fit), col = "blue")

Huang reduced growth model

Description

HuangRM function to fit the Huang reduced growth model to a reduced microbial growth curve. Returns the model parameters estimated according to data collected in microbial growth experiments.

Usage

HuangRM(t, Y0, MUmax, lag)

Arguments

Details

Model's inputs are:

t: time, assuming time zero as the beginning of the experiment.

Y(t): the natural logarithm of the microbial concentration (ln(N(t))) measured at time t.

Users should make sure that the microbial concentration input is entered in natural logarithm, Y(t) = ln(N(t)).

Value

A numeric vector with the fitted values

Author(s)

Vasco Cadavez, vcadavez@ipb.pt & Ursula Gonzales-Barron, ubarron@ipb.pt

References

Huang L (2008). “Growth Kinetics of Listeria monocytogenes in Broth and Beef Frankfurters-Determination of Lag Phase Duration and Exponential Growth Rate under Isothermal Conditions.” Journal of Food Science, 73(5), E235-E242. doi:10.1111/j.1750-3841.2008.00785.x.

Examples

## Example: Huang reduced model
library(gslnls)
data(growthred) # simulated data set.
initial_values <- list(Y0 = 0, MUmax = 1.7, lag = 5) # define the initial values
fit <- gsl_nls(lnN ~ HuangRM(Time, Y0, MUmax, lag),
  data = growthred,
  start = initial_values
)
summary(fit)

Richards no lag growth model

Description

RichardsNLM function to fit the Richards no lag growth model to an incomplete microbial growth curve. Returns the model parameters estimated according to data collected in microbial growth experiments.

Usage

RichardsNLM(t, Y0, Ymax, MUmax, m = 1)

Arguments

Details

Model's inputs are:

t: time, assuming time zero as the beginning of the experiment.

Y(t): the natural logarithm of the microbial concentration (ln(N(t))) measured at time t.

Users should make sure that the microbial concentration input is entered in natural logarithm, Y(t) = ln(N(t)).

Value

A numeric vector with the fitted values

Author(s)

Vasco Cadavez, vcadavez@ipb.pt and Ursula Gonzales-Barron, ubarron@ipb.pt

References

Richards JF (1959). “A flexible growth function for empirical use.” J Exp Bot, 1(10), 290-310.

Examples

## Example: Richards no lag model
library(gslnls)
data(growthnolag) # simulated data set.
initial_values <- list(Y0 = 0, Ymax = 22, MUmax = 1.7)
## Fitting the function to the experimental data
fit <- gsl_nls(lnN ~ RichardsNLM(Time, Y0, Ymax, MUmax),
  data = growthnolag,
  start = initial_values
)
summary(fit)

Rosso full growth model

Description

RossoFM function to fit the Rosso full growth model to complete microbial growth curve. Returns the model parameters estimated according to data collected in microbial growth experiments.

Usage

RossoFM(t, Y0, Ymax, MUmax, lag)

Arguments

Details

Model's inputs are:

t: time, assuming time zero as the beginning of the experiment.

Y(t): the natural logarithm of the microbial concentration (ln(N(t))) measured at time t.

Users should make sure that the microbial concentration input is entered in natural logarithm, Y(t) = ln(X(t)).

Value

A numeric vector with the fitted values

Author(s)

Vasco Cadavez (vcadavez@ipb.pt) and Ursula Gonzales-Barron (ubarron@ipb.pt)

References

Rosso L, Bajard S, Flandrois JP, Lahellec C, Fournaud J, Veit P (1996). “Differential growth of Listeria monocytogenes at 4 and 8 ºC: Consequences for the Shelf Life of Chilled Products.” Journal of Food Protection, 59(9), 944-949. ISSN 0362-028X, doi:10.4315/0362-028X-59.9.944.

Examples

## Example: Rosso full model
library(gslnls)
data(growthfull) # simulated data set.
initial_values <- list(Y0 = 0.04, Ymax = 21, MUmax = 1.9, lag = 5.0) # define the initial values
fit <- gsl_nls(lnN ~ RossoFM(Time, Y0, Ymax, MUmax, lag),
  data = growthfull,
  start = initial_values
)
summary(fit)

Weibull inactivation model Mafart

Description

WeibullM inactivation model for microbial inactivation curve. Returns the model parameters estimated according to data collected in microbial inactivation experiments.

Usage

WeibullM(x, Y0, sigma, alpha)

Arguments

Details

The model's inputs are:

t: time, assuming time zero as the beginning of the experiment.

Y(t): the natural logarithm of the bacterial concentration (ln(N(t))) measured at time t.

Users should make sure that the bacterial concentration input is entered in natural logarithm, Y(t) = ln(N(t)).

Value

A numeric vector with the fitted values

Author(s)

Vasco Cadavez vcadavez@ipb.pt and Ursula Gonzales-Barron ubarron@ipb.pt

References

Mafart P, Couvert O, Gaillard S, Leguerinel (2002). “On calculating sterility in thermal preservation methods: application of the Weibull frequency distribution model.” International Journal of Food Microbiology, 72, 107-113.

Examples

library(gslnls)
data(bixina)
initial_values <- list(Y0 = 5.75, sigma = 12.8, alpha = 2.4)
fit <- gsl_nls(lnN ~ WeibullM(Time, Y0, sigma, alpha),
  data = bixina,
  start = initial_values
)
summary(fit)

plot(lnN ~ Time, data = bixina)
lines(bixina$Time, predict(fit), col = "blue")

Weibull inactivation modified model Mafart

Description

WeibullMM inactivation model for microbial inactivation curve. Returns the model parameters estimated according to data collected in microbial inactivation experiments.

Usage

WeibullMM(x, Y0, Yres, sigma, alpha)

Arguments

Details

The model's inputs are:

t: time, assuming time zero as the beginning of the experiment.

Y(t): the bacterial concentration ($Y(t)$) measured at time t.

Users should make sure to use the base 10 logarithm bacterial concentration (Y(t)) as input.

Value

A numeric vector with the fitted values

Author(s)

Vasco Cadavez vcadavez@ipb.pt and Ursula Gonzales-Barron ubarron@ipb.pt

References

Mafart et al. (2005).

Examples

library(gslnls)
data(bixina)
initial_values <- list(Y0 = 5.6, Yres = 1, sigma = 2, alpha = 1)
bixina$logN <- log10(exp(bixina$lnN))
fit <- gsl_nls(logN ~ WeibullMM(Time, Y0, Yres, sigma, alpha),
  data = bixina,
  start = initial_values
)
summary(fit)

plot(logN ~ Time, data = bixina)
lines(bixina$Time, predict(fit), col = "blue")

Weibull inactivation model Peleg and Huang

Description

WeibullPH inactivation model for microbial inactivation curve. Returns the model parameters estimated according to data collected in microbial inactivation experiments.

Usage

WeibullPH(t, Y0, k, alpha)

Arguments

Details

The model's inputs are:

t: time, assuming time zero as the beginning of the experiment.

Y(t): the natural logarithm of the bacterial concentration X(t) measured at time t.

Users should make sure that the bacterial concentration input is entered in natural logarithm, Y(t) = ln(X(t)).

The following parameters can be estimated using Weibull function:

t: is heating time under a constant temperature

Y0: is the initial (time=0) bacterial counts in natural logarithm of the initial bacterial counts;

k: is the inactivation rate (ln units/time)

alpha: is the shape parameter of the survival curve

Value

A numeric vector with the fitted values

Author(s)

Vasco Cadavez vcadavez@ipb.pt and Ursula Gonzales-Barron ubarron@ipb.pt

References

Huang L (2009). “Thermal inactivation of Listeria monocytogenes in ground beef under isothermal and dynamic temperature conditions.” Journal of Food Engineering, 90(3), 380-387. ISSN 0260-8774, doi:10.1016/j.jfoodeng.2008.07.011.

Examples

library(gslnls)
data(bixina)
initial_values <- list(Y0 = 6.0, k = 1.0, alpha = 0.2)
fit <- gsl_nls(lnN ~ WeibullPH(Time, Y0, k, alpha),
  data = bixina,
  start = initial_values
)
summary(fit)

plot(lnN ~ Time, data = bixina)
lines(bixina$Time, predict(fit), col = "blue")

Zwietering full growth model

Description

ZwieteringFM function to fit the Zwietering full growth model to a complete microbial growth curve. Returns the model parameters estimated according to data collected in microbial growth experiments.

Usage

ZwieteringFM(t, Y0, Ymax, MUmax, lag)

Arguments

Details

Model's inputs are:

t: time, assuming time zero as the beginning of the experiment.

Y(t): the natural logarithm of the microbial concentration (ln(N(t))) measured at time t.

Users should make sure that the microbial concentration input is entered in natural logarithm, Y(t) = ln(X(t)).

Value

A numeric vector with the fitted values

Author(s)

Vasco Cadavez vcadavez@ipb.pt and Ursula Gonzales-Barron ubarron@ipb.pt

References

Zwietering MH, Jongenburger I, Rombouts FM, van't Riet K (1990). “Modeling of the Bacterial Growth Curve.” Applied and Environmental Microbiology, 56(6), 1875-1881. ISSN 0099-2240.

Examples

## Example: Zwietering full model
library(gslnls)
data(growthfull) # simulated data set.
initial_values <- list(Y0 = 0, Ymax = 22, MUmax = 1.7, lag = 5) # define the initial values
fit <- gsl_nls(lnN ~ ZwieteringFM(Time, Y0, Ymax, MUmax, lag),
  data = growthfull,
  start = initial_values
)
summary(fit)

Data of aw

Description

A dataset containing water activity and growth rate data

Usage

data(aw)

Format

A data frame with 9 rows and 3 variables:

aw

Water activity

sqrtGR

Square root of the growth rate

GR

Growth rate

Bias and accuracy factors

Description

bias_factor() and accuracy_factor() calculate the multiplicative model bias and accuracy factors commonly used in predictive microbiology model validation.

Usage

bias_factor(observed, predicted)

accuracy_factor(observed, predicted)

Arguments

Value

A numeric scalar.

Examples

observed <- c(7.0, 6.0, 5.0)
predicted <- c(7.1, 5.9, 5.2)
bias_factor(observed, predicted)
accuracy_factor(observed, predicted)

Data concerning Staphylococcus aureus microbial inactivation in beef

Description

A dataset containing time (minutes), repetition and number of microrganisms (ln N).

Usage

data(bixina)

Format

A data frame with 18 rows and 3 variables:

Time

Time in minutes

Rep

Repetition

lnN

Number of microrganism in ln scale

Compare fitted predmicror models

Description

compare_models() combines the output of fit_metrics() for two or more fitted models. It is useful when choosing between alternative primary growth, inactivation, or cardinal models fitted to the same response scale.

Usage

compare_models(..., sort_by = c("AIC", "BIC", "RMSE", "MAE", "none"))

Arguments

Value

A data frame with one row per fitted model.

Examples

data(growthfull)
huang <- fit_growth(
  growthfull,
  model = "HuangFM",
  time = "Time",
  response = "lnN",
  start = list(Y0 = 0, Ymax = 22, MUmax = 1.7, lag = 5)
)
baranyi <- fit_growth(
  growthfull,
  model = "BaranyiFM",
  time = "Time",
  response = "lnN",
  start = list(Y0 = 0, Ymax = 22, MUmax = 1.7, lag = 5)
)
compare_models(huang = huang, baranyi = baranyi)

Create a dynamic environmental profile

Description

dynamic_profile() stores time-varying environmental conditions such as temperature, pH, or water activity for dynamic predictive microbiology simulations. The profile is interpolated internally when dynamic models are solved between observed profile points.

Usage

dynamic_profile(time, temperature = NULL, ph = NULL, aw = NULL, ...)

Arguments

Value

A predmicror_dynamic_profile data frame sorted by time.

See Also

predict_dynamic_growth(), predict_dynamic_inactivation()

Examples

profile <- dynamic_profile(
  time = c(0, 5, 10, 15),
  temperature = c(10, 4, 12, 20)
)
profile

Finite-difference sensitivity for dynamic predictions

Description

dynamic_sensitivity() perturbs parameters in a dynamic prediction and returns unscaled and scaled sensitivity coefficients. It is designed as a lightweight diagnostic for sampling design and parameter identifiability.

Usage

dynamic_sensitivity(
  type = c("growth", "inactivation"),
  profile,
  start,
  parameters = NULL,
  relative_delta = 1e-06,
  times = NULL,
  ...
)

Arguments

Value

A data frame with time, parameter, prediction, sensitivity, and scaled sensitivity.

Examples

profile <- dynamic_profile(time = c(0, 5, 10), temperature = c(10, 15, 20))
sens <- dynamic_sensitivity(
  "growth",
  profile = profile,
  start = list(logN0 = 2, logNmax = 8, a = 0.08, Tmin = 7, lag = 1),
  times = seq(0, 10, by = 2),
  dt = 0.25
)
head(sens)

Fit a cardinal parameter model

Description

fit_cardinal() validates the input data, builds the nonlinear model formula, fits it with gslnls::gsl_nls(), and returns a predmicror_fit object.

Usage

fit_cardinal(data, model, x, response = "sqrtGR", start, ...)

Arguments

Details

Cardinal models expect the response to be the square root of the growth rate, usually sqrtGR.

Value

A predmicror_fit object with the fitted model and metadata.

Examples

data(salmonella)
fit <- fit_cardinal(
  salmonella,
  model = "CMTI",
  x = "Temp",
  response = "sqrtGR",
  start = list(Tmax = 42, Tmin = 1, MUopt = 1, Topt = 37)
)
coef(fit)

Fit dynamic microbial growth models

Description

fit_dynamic_growth() estimates selected parameters of a dynamic Huang-type growth model by repeatedly solving the dynamic model and minimizing the residual sum of squares against observed data.

Usage

fit_dynamic_growth(
  data,
  profile,
  time,
  response,
  start,
  estimate = NULL,
  fixed = NULL,
  lower = NULL,
  upper = NULL,
  model = "huang",
  secondary = "huang_sqrt",
  scale = c("log10", "ln"),
  dt = 0.01,
  method = "rk4",
  temperature = "temperature",
  optimizer = "L-BFGS-B",
  control = list(),
  ...
)

Arguments

Value

A predmicror_dynamic_fit object.

Examples

profile <- dynamic_profile(time = c(0, 10), temperature = c(20, 20))
obs <- data.frame(time = c(0, 5, 10), logN = c(2, 3.3, 5.1))
fit <- fit_dynamic_growth(
  obs,
  profile = profile,
  time = "time",
  response = "logN",
  start = list(logN0 = 2, logNmax = 8, MUmax = 0.4, lag = 0),
  estimate = "MUmax",
  secondary = "constant",
  dt = 0.25
)
coef(fit)

Fit dynamic microbial inactivation models

Description

fit_dynamic_inactivation() estimates selected parameters of a dynamic Weibull-Peleg inactivation model by minimizing the residual sum of squares against observed data.

Usage

fit_dynamic_inactivation(
  data,
  profile,
  time,
  response,
  start,
  estimate = NULL,
  fixed = NULL,
  lower = NULL,
  upper = NULL,
  model = "weibull_peleg",
  secondary = "constant",
  dt = 0.01,
  method = "rk4",
  temperature = "temperature",
  optimizer = "L-BFGS-B",
  control = list(),
  ...
)

Arguments

Value

A predmicror_dynamic_fit object.

Examples

profile <- dynamic_profile(time = c(0, 10), temperature = c(60, 60))
obs <- data.frame(time = c(0, 5, 10), logN = c(7, 6, 5))
fit <- fit_dynamic_inactivation(
  obs,
  profile = profile,
  time = "time",
  response = "logN",
  start = list(logN0 = 7, b = 0.15, n = 1),
  estimate = "b",
  dt = 0.25
)
coef(fit)

Fit a primary growth model

Description

fit_growth() validates the input data, builds the nonlinear model formula, fits it with gslnls::gsl_nls(), and returns a predmicror_fit object.

Usage

fit_growth(data, model, time, response = "lnN", start, ...)

Arguments

Details

Growth models expect the response to be the natural logarithm of the microbial concentration, usually lnN.

Value

A predmicror_fit object with the fitted model and metadata.

Examples

data(growthfull)
fit <- fit_growth(
  growthfull,
  model = "HuangFM",
  time = "Time",
  response = "lnN",
  start = list(Y0 = 0, Ymax = 22, MUmax = 1.7, lag = 5)
)
coef(fit)

Fit a microbial inactivation model

Description

fit_inactivation() validates the input data, builds the nonlinear model formula, fits it with gslnls::gsl_nls(), and returns a predmicror_fit object.

Usage

fit_inactivation(data, model, time, response = "logN", start, ...)

Arguments

Details

Inactivation models expect the response to be the base 10 logarithm of the microbial concentration, usually logN.

Value

A predmicror_fit object with the fitted model and metadata.

Examples

data(mafart2005Li11)
fit <- fit_inactivation(
  mafart2005Li11,
  model = "WeibullM",
  time = "Time",
  response = "logN",
  start = list(Y0 = 10, sigma = 3, alpha = 1)
)
coef(fit)

Calculate model diagnostics for a fitted predmicror model

Description

fit_metrics() summarizes goodness-of-fit and information criteria for a predmicror_fit object. The metrics are calculated on the response scale used in the fitted model.

Usage

fit_metrics(object, ...)

## Default S3 method:
fit_metrics(object, ...)

## S3 method for class 'predmicror_fit'
fit_metrics(object, ...)

Arguments

Value

A one-row data frame with columns: fit, model, type, response, response_scale, n (observations), p (parameters), SSE, RMSE, MAE, bias, RSE, R2, adj_R2, logLik, AIC, BIC, and converged.

Examples

data(growthfull)
fit <- fit_growth(
  data = growthfull,
  model = "HuangFM",
  time = "Time",
  response = "lnN",
  start = list(Y0 = 0, Ymax = 22, MUmax = 1.7, lag = 5)
)
fit_metrics(fit)

Fit omnibus predictive microbiology models

Description

fit_omnibus() fits a nonlinear mixed-effects model in which the primary model is one of the parameterised predmicror primary models and one or more primary-model parameters are described by secondary covariate formulas.

Usage

fit_omnibus(
  data,
  type = c("growth", "inactivation"),
  primary,
  time,
  response,
  group,
  secondary = NULL,
  random,
  correlation = NULL,
  start,
  method = "ML",
  control = NULL,
  ...
)

fit_omnibus_growth(
  data,
  primary,
  time,
  response,
  group,
  secondary = NULL,
  random,
  correlation = NULL,
  start,
  method = "ML",
  control = NULL,
  ...
)

fit_omnibus_inactivation(
  data,
  primary,
  time,
  response,
  group,
  secondary = NULL,
  random,
  correlation = NULL,
  start,
  method = "ML",
  control = NULL,
  ...
)

Arguments

Value

A predmicror_omnibus_fit object.

Examples

set.seed(1)
dat <- do.call(rbind, lapply(1:4, function(g) {
  Time <- c(1, 2, 4, 6, 8, 10)
  sigma <- 5 + 0.4 * g
  data.frame(
    Condition = g,
    Time = Time,
    Temp = 55 + g,
    logN = WeibullM(Time, Y0 = 7, sigma = sigma, alpha = 1.1) +
      rnorm(length(Time), 0, 0.03)
  )
}))
fit <- fit_omnibus_inactivation(
  dat,
  primary = "WeibullM",
  time = "Time",
  response = "logN",
  group = "Condition",
  secondary = list(sigma = ~ Temp),
  random = Y0 ~ 1,
  start = c(Y0 = 7, sigma = 1, sigma.Temp = 0.08, alpha = 1)
)
fit_metrics(fit)

Data of a complete curve of microbial growth

Description

A dataset containing simulated data for a full growth model

Usage

data(growthfull)

Format

A data frame with 13 rows and 3 variables.

Time

Time in minutes

logN

Number of microrganism in log10 scale

lnN

Number of microrganism in ln scale

Data of a no lag curve of microbial growth

Description

A dataset containing simulated data for a no lag growth model

Usage

data(growthnolag)

Format

A data frame with 10 rows and 3 variables.

Time

Time in minutes

logN

Number of microrganism in log10 scale

lnN

Number of microrganism in ln scale

Data of a reduced curve of microbial growth

Description

A dataset containing simulated data for a reduced growth model

Usage

data(growthred)

Format

A data frame with 9 rows and 3 variables.

Time

Time in minutes

logN

Number of microrganism in log10 scale

lnN

Number of microrganism in ln scale

Data of INH antimicrobials

Description

A dataset containing antimicrobial concentration and growth rate data

Usage

data(inh)

Format

A data frame with 8 rows and 3 variables.

Conce

Antimicrobial concentration

sqrtGR

Square root of the growth rate

GR

Growth rate

Data of microbial inactivation Albert and Mafart (2005)

Description

A dataset containing time and log10 counts for microbial inactivation

Usage

data(mafart2005Li11)

Format

A data frame with 10 rows and 2 variables.

Time

Time

logN

Number of microrganism in log10 scale

Define an omnibus secondary model

Description

omnibus_secondary() defines a secondary model specification that can be used inside fit_omnibus(), fit_omnibus_growth(), or fit_omnibus_inactivation(). It allows a primary-model parameter to be described by another parameterised model already available in predmicror.

Usage

omnibus_secondary(model, x, transform = c("identity", "square"))

Arguments

Value

An omnibus_secondary specification.

Examples

omnibus_secondary("CMTI", x = "Temp")
omnibus_secondary("CMTI", x = "Temp", transform = "square")

Data pH

Description

A dataset containing pH and growth rate data

Usage

data(ph)

Format

A data frame with 14 rows and 3 variables.

pH

pH of food

sqrtGR

Square root of the growth rate

GR

Growth rate

Predict microbial growth under dynamic environmental conditions

Description

predict_dynamic_growth() solves a differential Huang-type growth model over a time-varying environmental profile. It is intended for forward prediction under dynamic temperature conditions and uses an internal fourth-order Runge-Kutta solver.

Usage

predict_dynamic_growth(
  profile,
  model = "huang",
  secondary = "huang_sqrt",
  start,
  times = NULL,
  scale = c("log10", "ln"),
  dt = 0.01,
  method = "rk4",
  temperature = "temperature"
)

Arguments

Value

A predmicror_dynamic_prediction data frame with time, prediction, lnN, logN, temperature, and metadata attributes.

Examples

profile <- dynamic_profile(
  time = c(0, 5, 10, 15, 20),
  temperature = c(10, 4, 15, 15, 10)
)
pred <- predict_dynamic_growth(
  profile = profile,
  start = list(logN0 = 2, logNmax = 8.8, a = 0.0886, Tmin = 8.91, lag = 2),
  dt = 0.25
)
head(pred)

Predict microbial inactivation under dynamic environmental conditions

Description

predict_dynamic_inactivation() solves a dynamic Weibull-Peleg type inactivation model over a time-varying environmental profile. The output is returned on the base-10 logarithmic scale.

Usage

predict_dynamic_inactivation(
  profile,
  model = "weibull_peleg",
  secondary = "constant",
  start,
  times = NULL,
  dt = 0.01,
  method = "rk4",
  temperature = "temperature"
)

Arguments

Value

A predmicror_dynamic_prediction data frame with time, logN, log_survival, temperature, and metadata attributes.

Examples

profile <- dynamic_profile(time = c(0, 10, 20), temperature = c(55, 58, 60))
pred <- predict_dynamic_inactivation(
  profile = profile,
  start = list(logN0 = 7, b_ref = 0.15, T_ref = 55, z = 8, n = 1.2),
  secondary = "z_value",
  dt = 0.25
)
head(pred)

Assistant for predmicror

Description

Answer questions about predmicror models using a deterministic local registry and, when available, an optional local Ollama model for prose. Model-fitting code is generated from package metadata and statically checked before being returned.

Usage

predmicror_assistant(
  query,
  model = "llama3-groq-tool-use:8b",
  root = NULL,
  data = NULL,
  file = NULL,
  sheet = NULL,
  sep = NULL,
  dec = ".",
  na.strings = c("", "NA"),
  task = NULL,
  time = NULL,
  response = NULL,
  temperature = NULL,
  ph = NULL,
  aw = NULL,
  inhibitor = NULL,
  return_context = FALSE,
  conversation = NULL,
  backend = c("auto", "ollama", "deterministic"),
  prefer_wrappers = TRUE,
  verify_code = TRUE,
  return_trace = FALSE
)

Arguments

Value

Character response by default; list with answer and context when return_context = TRUE or list with trace when return_trace = TRUE.

Examples

predmicror_assistant("How do I fit a Huang model?", backend = "deterministic")

Launch the predmicror assistant Shiny app

Description

Starts the local assistant app bundled with the package. The app can read delimited text and Excel files, preview uploaded data, profile columns, apply optional manual column overrides, and pass the data profile to predmicror_assistant(). If the bundled app is not available, a fallback app is created from the installed R functions.

Usage

predmicror_assistant_app(
  model = "llama3-groq-tool-use:8b",
  root = NULL,
  host = "127.0.0.1",
  port = NULL,
  launch.browser = interactive()
)

Arguments

Value

Runs the Shiny app.

Examples

predmicror_assistant_app()

Extract fitted values and residuals from a predmicror fit

Description

predmicror_augment() returns the original data, or optional new data, with columns containing fitted values and residuals. It is intentionally lightweight and does not require the broom package.

Usage

predmicror_augment(object, ...)

## Default S3 method:
predmicror_augment(object, ...)

## S3 method for class 'predmicror_fit'
predmicror_augment(object, newdata = NULL, ...)

## S3 method for class 'predmicror_fit'
as.data.frame(x, row.names = NULL, optional = FALSE, ...)

Arguments

Value

A data frame containing the original columns plus .fitted, .resid when the response column is available, .model, and .type.

Examples

data(growthfull)
fit <- fit_growth(
  data = growthfull,
  model = "HuangFM",
  time = "Time",
  response = "lnN",
  start = list(Y0 = 0, Ymax = 22, MUmax = 1.7, lag = 5)
)
head(predmicror_augment(fit))

Methods for predmicror_fit objects

Description

These methods delegate to the fitted nonlinear least-squares object returned by gslnls::gsl_nls(), while preserving the model metadata added by the ⁠fit_*()⁠ wrappers.

Usage

## S3 method for class 'predmicror_fit'
print(x, ...)

## S3 method for class 'predmicror_fit'
summary(object, ...)

## S3 method for class 'predmicror_fit'
predict(object, newdata = NULL, ...)

## S3 method for class 'predmicror_fit'
plot(
  x,
  xlab = x$x,
  ylab = paste0(x$response, " (", x$response_scale, ")"),
  ...
)

## S3 method for class 'predmicror_fit'
coef(object, ...)

## S3 method for class 'predmicror_fit'
fitted(object, ...)

## S3 method for class 'predmicror_fit'
residuals(object, ...)

## S3 method for class 'predmicror_fit'
vcov(object, ...)

## S3 method for class 'predmicror_fit'
logLik(object, ...)

## S3 method for class 'predmicror_fit'
AIC(object, ..., k = 2)

## S3 method for class 'predmicror_fit'
BIC(object, ...)

Arguments

Value

The value returned by the corresponding method for the underlying nonlinear model object. plot() invisibly returns x.

List models available through the fitting wrappers

Description

List models available through the fitting wrappers

Usage

predmicror_models(type = "all")

Arguments

Value

A data frame with the model type, model name, expected response scale, and parameters that must be supplied in start.

Examples

predmicror_models()
predmicror_models("growth")

Methods for omnibus fits

Description

Methods for omnibus fits

Usage

## S3 method for class 'predmicror_omnibus_fit'
print(x, ...)

## S3 method for class 'predmicror_omnibus_fit'
summary(object, ...)

## S3 method for class 'predmicror_omnibus_fit'
coef(object, ...)

## S3 method for class 'predmicror_omnibus_fit'
fitted(object, ...)

## S3 method for class 'predmicror_omnibus_fit'
residuals(object, ...)

## S3 method for class 'predmicror_omnibus_fit'
predict(object, newdata = NULL, level = 0, ...)

## S3 method for class 'predmicror_omnibus_fit'
logLik(object, ...)

## S3 method for class 'predmicror_omnibus_fit'
AIC(object, ..., k = 2)

## S3 method for class 'predmicror_omnibus_fit'
BIC(object, ...)

Arguments

Value

The value returned by the corresponding method.

Potential growth of Salmonella typhimurium on cooked chicken

Description

A dataset containing the specific growth rate of Salmonella typhimurium on cooked chicken

Usage

data(salmonella)

Format

A data frame with 21 rows and 3 variables:

Temp

Temperature (ºC)

GR

Specific growth rate (log10/h)

sqrtGR

Square root of the specific growth rate (GR)

References

Oscar TP (2002). “Development and validation of a tertiary simulation model for predicting the potential growth of Salmonella typhimurium on cooked chicken.” International Journal of Food Microbiology, 76(3), 177-190. ISSN 0168-1605, doi:10.1016/S0168-1605(02)00025-9.

Validate an omnibus fit by leaving out one group

Description

Validate an omnibus fit by leaving out one group

Usage

validate_omnibus_leave_one_out(object, group_value, level = 0, ...)

Arguments

Value

A list with the refitted model, validation data, predictions, residuals, bias factor, and accuracy factor.