fitOneExp: Fit multiple dose response models for a single drug in a...

Description Usage Arguments Value Examples

Description

The dose response mode is usually fitted with original dose (not log10 transformed). The computed ICx values however, can be in either log10 scale or the original scale.

Usage

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
fitOneExp(
  dat,
  drug = NA,
  cellLine = NA,
  unit = NA,
  models = c("LL.3", "LL.3u", "sigEmax", "logistic", "linlog"),
  cols = NA,
  log.d = TRUE,
  percentile = seq(0.1, 0.9, by = 0.1),
  alpha = 0.01,
  fitCtr = FALSE,
  standardize = TRUE,
  interpolation = TRUE,
  plot = FALSE,
  transparency = 1,
  ...
)

Arguments

dat

a 2-column data frame with first column as dose and second column response. Controls are decided by dose = 0.

drug

drug for this analysis.

cellLine

cell line for this analysis.

unit

unit of drug concentration.

models

models the user may specify.

cols

colors of dose-response curves.

log.d

in computed ICx values, whether to return the log10 dose or the dose without this transformation.

percentile

IC percentile.

alpha

outlier significance level.

fitCtr

whether the model is fitted with control data.

standardize

whether to standardize (scale) the data based on control points. This should be disabled when no control data is supplied.

interpolation

whether calculate ICx through interpolation or through root finding. This parameter is passed to computeIC() function. When interpolation = TRUE, ICx value will be bounded by observed minimum and maximum dose; otherwise, ICx is calculated through root finding and thus can be outside this boundary (extrapolated).

plot

whether to draw the dose response curve.

transparency

a value between 0 and 1 specifying transparency through alpha blending; transparency = 0 means totally transparent.

...

additional parameters to plotOneExp().

Value

a list containing elements with:

fits,

models,

cols,

ICmat, IC matrix from all specified models as well as RSE and model name

ICx, IC values from the best model

datWithOutlierStatus the input data with outlier status appended

bestModel the best model by RSE

RSEs

Examples

1
2
3
fitExp <- fitOneExp(ryegrass[, c(2, 1)], drug = '', cellLine = '', unit = '',
                    models = c('sigEmax', 'LL.4', 'LL.5', 'LL.3', 'LL.3u', 'logistic'),
                    alpha = 0.05, interpolation = TRUE) 

lshen1/drexplorer2 documentation built on June 2, 2020, 9:27 p.m.