At very high concentrations the measured signal can decrease (hook effect). This function compresses the post-peak delta toward the peak value.
Usage
correct_prozone(
stdframe,
prop_diff = 0.1,
dil_scale = 2,
response_variable = "mfi",
independent_variable = "concentration",
include_col = "included",
verbose = FALSE
)Arguments
- stdframe
Data frame of standard curve data.
- prop_diff
Numeric. Dampening factor (e.g. 0.1).
- dil_scale
Numeric. Dilution scale factor (e.g. 2).
- response_variable
Character. Response column name.
- independent_variable
Character. Concentration column name.
- include_col
Character. Logical column marking fitted rows. Absent = all rows included (backward compatible).
- verbose
Logical.
Value
stdframe (all rows) with post-peak response values adjusted. The
peak reference is attached as attr(., "prozone_peak_response") and
attr(., "prozone_logc_at_peak").
Details
The peak (max_response and logc_at_max) is a set-level statistic
computed from the included points only. The post-peak reflection is then
applied to all rows relative to that peak, so masked points beyond the
hook are dampened onto the same reference as the fitted points. Rows are
never dropped (grain is preserved); rows with a missing response or
concentration are passed through untouched.