Skip to contents

SEC/GPC Baseline Correction

Source: R/step_sec_baseline.R
step_sec_baseline.Rd

step_sec_baseline() creates a specification of a recipe step that applies baseline correction optimized for Gel Permeation Chromatography (GPC) or Size Exclusion Chromatography (SEC) data. This method estimates the baseline by interpolating between baseline regions at the start and end of the chromatogram.

Usage

step_sec_baseline(
  recipe,
  measures = NULL,
  left_frac = 0.05,
  right_frac = 0.05,
  method = "linear",
  rf_span = 2/3,
  rf_maxit = c(20, 20),
  role = NA,
  trained = FALSE,
  skip = FALSE,
  id = recipes::rand_id("sec_baseline")
)

Arguments

recipe

A recipe object. The step will be added to the sequence of operations for this recipe.

measures

An optional character vector of measure column names to process. If NULL (the default), all measure columns (columns with class measure_list) will be processed.

left_frac

Fraction of points from the beginning to use as the left baseline region. Default is 0.05 (first 5% of data points). Not used for method = "rf".

right_frac

Fraction of points from the end to use as the right baseline region. Default is 0.05 (last 5% of data points). Not used for method = "rf".

method

Method for baseline estimation. One of:

  • "linear" (default): Linear interpolation between left and right means

  • "median": Uses median of baseline regions (more robust to outliers)

  • "spline": Smooth spline through baseline regions

  • "rf": Robust local regression (IDPmisc::rfbaseline). Best for complex baselines with curvature or drift. Does not use left_frac/right_frac.

rf_span

Span parameter for RF baseline (only used when method = "rf"). A numeric value between 0 and 1 specifying the fraction of points used for local regression. Default is 2/3. Higher values produce smoother baselines.

rf_maxit

Maximum iterations for RF baseline (only used when method = "rf"). A numeric vector of length 2 specifying max iterations for the two stages of the robust fitting. Default is c(20, 20).

role

Not used by this step since no new variables are created.

trained

A logical to indicate if the quantities for preprocessing have been estimated.

skip

A logical. Should the step be skipped when the recipe is baked?

id

A character string that is unique to this step to identify it.

Value

An updated version of recipe with the new step added to the sequence of any existing operations.

Details

GPC/SEC chromatograms typically have distinct baseline regions at the beginning and end where no polymer elutes. This step leverages this characteristic by:

  1. Identifying baseline regions at the start and end of the chromatogram

  2. Computing a representative baseline value for each region (mean or median)

  3. Interpolating between these values to estimate the full baseline

  4. Subtracting the estimated baseline from the signal

The left_frac and right_frac parameters control how much of the chromatogram is considered "baseline". Choose values that:

  • Include only the flat, signal-free regions

  • Exclude any polymer peaks or system peaks

  • Are large enough to average out noise

Unlike general-purpose baseline methods like ALS or polynomial fitting, this approach is specifically designed for the characteristic shape of GPC/SEC chromatograms and is computationally very fast.

No selectors should be supplied to this step function. The data should be in the internal format produced by measure::step_measure_input_wide() or measure::step_measure_input_long().

Tidying

When you tidy() this step, a tibble with columns terms, left_frac, right_frac, method, and id is returned.

See also

measure::step_measure_baseline_als() for general-purpose baseline correction, measure::step_measure_detrend() for simple trend removal.

Other sec-chromatography: step_sec_band_broadening(), step_sec_detector_delay(), step_sec_exclude_regions(), step_sec_integration_window(), step_sec_mw_averages(), step_sec_mw_distribution(), step_sec_mw_fractions(), step_sec_peaks_deconvolve(), step_sec_peaks_detect(), step_sec_peaks_refine()

Examples

if (FALSE) { # \dontrun{
library(recipes)
library(measure)

# SEC baseline correction with default settings
rec <- recipe(~., data = sec_triple_detect) |>
  step_measure_input_wide(starts_with("signal_")) |>
  step_sec_baseline() |>
  prep()

# Using median method for robustness to outliers
rec <- recipe(~., data = sec_triple_detect) |>
  step_measure_input_wide(starts_with("signal_")) |>
  step_sec_baseline(left_frac = 0.1, right_frac = 0.1, method = "median") |>
  prep()

# Using RF baseline for complex baseline shapes
rec <- recipe(~., data = sec_triple_detect) |>
  step_measure_input_wide(starts_with("signal_")) |>
  step_sec_baseline(method = "rf", rf_span = 0.5) |>
  prep()
} # }