Get Calibration Metrics — get_calibration

Extracts calibration fit metrics from a MidarExperiment object.

Usage

get_calibration_metrics(
  data = NULL,
  with_lod = TRUE,
  with_loq = TRUE,
  with_bias = TRUE,
  with_coefficients = TRUE,
  with_sigma = TRUE
)

data: A MidarExperiment object with QC metrics.
with_lod: Whether to include LoD in output. Default is TRUE.
with_loq: Whether to include LoQ in output. Default is TRUE.
with_bias: Whether to include bias in output. Default is TRUE.
with_coefficients: Whether to include regression coefficients. Default is TRUE.
with_sigma: Whether to include sigma in output. Default is TRUE.

A tibble with exported calibration metrics.

Requires prior computation of regression results using calc_calibration_results(). See its documentation for details.

feature_id: Feature identifier.
is_quantifier: Logical, indicates if the feature is a quantifier.
fit_model: Regression model used for fitting.
weighting: Weighting method used in fitting.
lowest_cal: Lowest nonzero calibration concentration.
highest_cal: Highest calibration concentration.
r.squared: R-squared value, indicating goodness of fit.
coef_a: Slope of the regression line (linear) or coefficient of the quadratic term (x^2) (quadratic).
coef_b: Intercept of the regression line (linear) or coefficient of the linear term (x) (quadratic).
coef_c: Intercept of the regression equation (quadratic). Set to NA for linear models.
sigma: Residual standard error of the model.
reg_failed: TRUE if regression fitting failed.
LoD = 3× the sample standard error of residuals / slope of the regression.
LoQ = 10× the sample standard error of residuals / slope of the regression.

Note: For LoD/LoQ calculations, the slope used in the formula is calculated at the lowest nonzero calibration point for quadratic fits.