Description Usage Arguments Details Value Methods (by class) See Also Examples
Fit a model and append it to the @model slot of the input object.
1 2 3 4 5 |
object |
object of class DRE |
... |
other arguments passed to the model-fitting function, e.g. to loess or nls |
type |
type of model to fit, character. Currently supports "melm", "loess" and "menls": see Details |
name |
a name to assign to the model, character |
loess.xlog |
fit response vs. ln(dose) when using loess |
control.weight |
weight to assing to the control.response when using loess with loess.xlog = F, numeric |
silent |
logical: should the report of error messages be suppressed? |
The following model types can be specified by using the type
argument:
melm
- fit a median effect equation model with parameters ic50 and m using linear least
squares lm
.
The data are linearized as log(fa/fu) = m*log(dose) - m*log(ic50),
where fu is fraction unaffected, fu = response/control.response. fa is fraction affected, fa = 1-fu.
If fu > 1, it is set to 0.9999
menls
- fit a median effect equation model with parameters ic50 and m using non-linear
least squares nls
loess
- fit a LOESS local regression model, using both dose-response data and the untreated control value,
i.e. object@control.response
.
It is recommended to assign a high weight to the control, if there is a relatively large number of untreated control
data points in an experiment. This is determined by the control.weight
argument.
Consider setting it to the ratio of the number of control data points to the number of replicates in the dose-response
data.
It is set to 10 by default.
An object of the same class as the input with a model object appended to its @model slot
DRE
: Fit a DRE object
This method fits a dose-response curve model and appends it to the model slot of a DRE object
1 2 3 4 | data(sim15_screen3)
x = fit(dre(sim15_screen3,"Cpd10"),type="melm")
model(x)$melm
plot(x)
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.