Skip to contents

Perform a biosensor model fit on response vs. concentration data of a single sample.

Source: R/dose-response-analysis.R
fl.drFitModel.Rd

fl.drFitModel fits the biosensor model proposed by Meyer et al. (2019) to the provided response (e.g., max_slope.spline vs. concentration data to determine the leakiness, sensitivity, induction fold-change, and cooperativity.

Usage

fl.drFitModel(conc, test, drID = "undefined", control = fl.control())

Arguments

conc

Vector of concentration values.

test

Vector of response parameter values of the same length as conc.

drID

(Character) The name of the analyzed condition

control

A fl.control object created with fl.control, defining relevant fitting options.

Value

A drFitFLModel object.

raw.conc

Raw data provided to the function as conc.

raw.test

Raw data for the response parameter provided to the function as test.

drID

(Character) Identifies the tested condition

fit.conc

Fitted concentration values.

fit.test

Fitted response values.

model

nls object generated by the nlsLM function.

parameters

List of parameters estimated from dose response curve fit.

  • yEC50: Response value related to EC50.

  • y.min: Minimum fluorescence ('leakiness', if lowest concentration is 0).

  • y.max: Maximum fluorescence.

  • fc: Fold change (y.max divided by y.min).

  • K: Concentration at half-maximal response ('sensitivity').

  • n: Cooperativity.

  • yEC50.orig: Response value for EC50 in original scale, if a transformation was applied.

  • K.orig: K in original scale, if a transformation was applied.

  • test.nm: Test identifier extracted from test.

fitFlag

(Logical) Indicates whether a spline could fitted successfully to data.

reliable

(Logical) Indicates whether the performed fit is reliable (to be set manually).

control

Object of class fl.control created with the call of fl.control.

Use plot.drFitModel to visualize the model fit.

References

Meyer, A.J., Segall-Shapiro, T.H., Glassey, E. et al. Escherichia coli “Marionette” strains with 12 highly optimized small-molecule sensors. Nat Chem Biol 15, 196–204 (2019). DOI: 10.1038/s41589-018-0168-3

Examples

# Create concentration values via a serial dilution
conc <- c(0, rev(unlist(lapply(1:18, function(x) 10*(2/3)^x))),10)

# Simulate response values via biosensor equation
response <- biosensor.eq(conc, y.min = 110, y.max = 6000, K = 0.5, n = 2) +
            0.01*6000*rnorm(10)

# Perform fit
TestRun <- fl.drFitModel(conc, response, drID = 'test', control = fl.control())
#> 
#> 
#> === Dose response curve estimation ================
#> --- EC 50 -----------------------------------------
#> --> test
#> sensitivity: 0.495 | yEC50: 3021 | fold change: 145.56 | leakiness: 41.2
#> 
#> 

print(summary(TestRun))
#>      yEC50    y.min    y.max       fc         K        n yEC50.orig    K.orig
#> 1 3020.854 41.22442 6000.484 145.5566 0.4948616 1.982843   3020.854 0.4948616
#>   test
#> 1   NA
plot(TestRun)