| Title: | Fitting Dose-Response Models Using an Evolutionary Algorithm |
| Version: | 0.1.4 |
| Description: | Fits dose-response models using an evolutionary algorithm to estimate the model parameters. The procedure currently can fit 3-parameter, 4-parameter, and 5-parameter log-logistic models as well as exponential models. Functions are also provided to plot, make predictions, and calculate confidence intervals for the resulting models. For details see "Nonlinear Dose-response Modeling of High-Throughput Screening Data Using an Evolutionary Algorithm", Ma, J., Bair, E., Motsinger-Reif, A.; Dose-Response 18(2):1559325820926734 (2020) <doi:10.1177/1559325820926734>. |
| Depends: | R (≥ 2.10.0) |
| License: | GPL-3 |
| Encoding: | UTF-8 |
| LazyData: | true |
| RoxygenNote: | 7.2.0 |
| NeedsCompilation: | no |
| Packaged: | 2022-05-28 09:29:10 UTC; ebair |
| Author: | Eric Bair [aut, cre], Jun Ma [aut] |
| Maintainer: | Eric Bair <eric.bair@sciome.com> |
| Repository: | CRAN |
| Date/Publication: | 2022-05-29 05:40:06 UTC |
Example dose response dataset
Description
A simple dose response dataset used to illustrate the use of the eadrm functions.
Usage
CarboA
Format
A list containing two elements:
- x
The drug concentration levels
- y
The response values
...
Calculates EC50 (or some other specified EC value)
Description
Calculates the concentration that induces a response corresponding to a specific proportion between the baseline and maximum. It is most commonly used to compute EC50.
Usage
calc.ec(eadrm.obj, ec.in = 0.5, ci.obj = NULL)
Arguments
eadrm.obj |
Fitted eadrm model object. |
ec.in |
A value between 0 and 1 corresponding to the desired proportion. Defaults to 0.5, in which case EC50 is computed. |
ci.obj |
Output from eadrm.ci that is used to compute confidence intervals. Defaults to NULL, in which case no confidence intervals are computed. |
Value
If no confidence intervals are requested, it returns the concentration corresponding to the requested proportion. Otherwise returns a list of three values ec.med, ec.l95, and ec.u95, which correspond to the median concentration and the corresponding lower/upper 95% confidence bounds.
See Also
Examples
ea.fit <- eadrm(CarboA$y, CarboA$x)
calc.ec(ea.fit)
Finds the dose that corresponds to a particular level of the response
Description
Calculates the concentration that induces a particular level of the response.
Usage
calc.ed(eadrm.obj, response, ci.obj = NULL)
Arguments
eadrm.obj |
Fitted eadrm model object. |
response |
The desired response level. |
ci.obj |
Output from eadrm.ci that is used to compute confidence intervals. Defaults to NULL, in which case no confidence intervals are computed. |
Value
If no confidence intervals are requested, it returns the concentration corresponding to the specified response. Otherwise returns a list of three values ed.med, ed.l95, and ed.u95, which correspond to the median concentration and the corresponding lower/upper 95% confidence bounds.
See Also
Examples
ea.fit <- eadrm(CarboA$y, CarboA$x)
calc.ec(ea.fit)
Fits a dose-response curve using an evolutionary algorithm
Description
Uses an evolutionary algorithm to fit a set of four possible dose-response models. The evolutionary parameters and stopping rules can be customized by the user.
Usage
eadrm(
obs,
xvals,
model = "h4",
pop.size = 1000,
stable.pop.size = 200,
num.tournaments = 20,
tournament.size = 10,
max.generations = 500,
stop.generations = 50,
epsilon = 0.1
)
Arguments
obs |
A vector of response values (y-values). |
xvals |
A vector of doses (x-values). |
model |
Type of dose-response model to fit. Possible values include "h3", "h4", and "h5" (corresponding to 3-parameter, 4-parameter, and 5-parameter log-logistic models, respectively), "e" (corresponding to an exponential model) and "all" (which allows the procedure to evaluate all four types of models). Defaults to "h4". |
pop.size |
The number of initial potential solutions. Defaults to 1000. |
stable.pop.size |
This quantity is divided by the number of tournaments to calculate the number of children in each generation. Defaults to 200. |
num.tournaments |
The number of tournaments in each generation. Defaults to 20. |
tournament.size |
The number of players (i.e., models to consider) in each tournament. Defaults to 10. |
max.generations |
The maximum number of generations. If this number is reached, the algorithm immediately terminates. Defaults to 500. |
stop.generations |
The algorithm will also terminate if there is no improvement in fitness in stop.generation generations. Defaults to 50. |
epsilon |
If three successive new models produce an improvement of less than epsilon in fitness, the procedure will terminate. Defaults to 0.1. |
Value
An object of class eadrm, which is a list containing the following elements:
- Model:
Specifies the type of model (i.e., "h3", "h4", "h5", or "e")
- R2:
Fitness for the final model
- params:
A vector of coefficients for the final model
- xvals:
The original x values (concentrations) for the model
- yvals:
The original y values (responses) for the model
Details
The procedure will initially generate pop.size possible solutions. The fitness for each solution will be calculated. In each generation, a series of tournaments are performed. The children of the surviving models from the previous generation are mutated. The model with the best fitness "wins" each tournament and survives to the next generation. The procedure continues until the maximum number of generations is reached or the fitness fails to improve substantially over a sufficient number of generations.
References
Ma, J., Bair, E., Motsinger-Reif, A. "Nonlinear Dose-response Modeling of High-Throughput Screening Data Using an Evolutionary Algorithm", Dose Response 18(2):1559325820926734 (2020).
Examples
ea.fit <- eadrm(CarboA$y, CarboA$x)
Computes confidence intervals for an eadrm model fit
Description
Calculates confidence intervals for an eadrm model fit by repeatedly fitting the model to the same data set and examining the distribution of the coefficients.
Usage
eadrm.ci(obs, xvals, model = "h4", ..., B = 1000)
Arguments
obs |
A vector of response values (y-values). |
xvals |
A vector of doses (x-values). |
model |
Type of dose-response model to fit. Possible values include "h3", "h4", and "h5" (corresponding to 3-parameter, 4-parameter, and 5-parameter log-logistic models, respectively) and "e" (corresponding to an exponential model). Defaults to "h4". |
... |
Additional parameters for the eadrm function. |
B |
Number of replicate models to fit. Defaults to 1000. |
Value
A list containing the following elements:
- med.est:
A vector of the median values of the coefficients across the B iterations
- l95.est,u95est:
Vectors of the lower/upper 95% confidence bounds for the coefficients across the B iterations
- replicate.mat:
A p x B matrix, where p is the number of coefficients in the model. Each column of B corresponds to the coefficients for one fitted model.
Details
This function calls the eadrm function B times with
the same parameters and records the model coefficients for each
iteration of the model. Confidence intervals for the coefficients
are calculated by examining the quantiles of the distribution
of the coefficients over the B iterations. A matrix of the
coefficients for each iteration is also calculated. This matrix
can be used to compute confidence intervals for predicted values
and estimates of EC50.
See Also
eadrm, predict.eadrm,
calc.ec, calc.ed
Examples
ea.ci <- eadrm.ci(CarboA$y, CarboA$x)
Plot an eadrm object
Description
Plots the data used to fit an eadrm object as well as the fitted dose-response curve
Usage
## S3 method for class 'eadrm'
plot(x, ...)
Arguments
x |
Object of class eadrm to plot. |
... |
Additional arguments to plot. Currently ignored. |
See Also
Examples
ea.fit <- eadrm(CarboA$y, CarboA$x)
plot(ea.fit)
Makes predictions from an eadrm object
Description
Similar to other predict methods, this function predicts fitted values (and optionally confidence intervals) for a eadrm object.
Usage
## S3 method for class 'eadrm'
predict(object, newx = object$xvals, ci.obj = NULL, ...)
Arguments
object |
Fitted model of class eadrm. |
newx |
Vector of new concentration values at which predictions are to be made. Defaults to the concentration values that were used to fit the model. |
ci.obj |
Output from eadrm.ci that is used to compute confidence intervals. Defaults to NULL, in which case no confidence intervals are computed. |
... |
Additional arguments passed to or from other methods. Currently ignored. |
Value
If no confidence intervals are requested, a vector of predicted responses for each concentration in newx is returned. Otherwise returns a list of three vectors yhat.med, yhat.l95, and yhat.u95, which correspond to the median predicted response and lower/upper 95% confidence bounds for each concentration in newx.
See Also
Examples
ea.fit <- eadrm(CarboA$y, CarboA$x)
predict(ea.fit)