Skip to contents

Matrix Effect Analysis

Source: R/matrix-effects.R
measure_matrix_effect.Rd

Quantifies matrix effects (ion suppression/enhancement) by comparing analyte response in matrix versus neat solution. This is essential for validating LC-MS/MS and other analytical methods where matrix interference is a concern.

Usage

measure_matrix_effect(
  data,
  response_col,
  sample_type_col,
  matrix_level,
  neat_level,
  concentration_col = NULL,
  analyte_col = NULL,
  group_cols = NULL,
  conf_level = 0.95
)

Arguments

data

A data frame containing response data.

response_col

Name of the column containing analyte responses.

sample_type_col

Name of the column indicating sample type (matrix vs neat/standard).

matrix_level

Value in sample_type_col indicating matrix samples.

neat_level

Value in sample_type_col indicating neat/standard samples.

concentration_col

Optional column for concentration levels. If provided, matrix effects are calculated per concentration.

analyte_col

Optional column for analyte names. If provided, matrix effects are calculated per analyte.

group_cols

Additional grouping columns (e.g., batch, matrix source).

conf_level

Confidence level for intervals. Default is 0.95.

Value

A measure_matrix_effect object containing:

  • results: Tibble with matrix effect percentages per group

  • statistics: Overall summary statistics

  • raw_data: Data used for calculations

Details

Matrix Effect Calculation

Matrix effect (ME%) is calculated as: ME% = (response_in_matrix / response_in_neat) * 100

Or equivalently: ME% = 100 + ((response_in_matrix - response_in_neat) / response_in_neat) * 100

Interpretation

  • ME = 100%: No matrix effect

  • ME > 100%: Ion enhancement

  • ME < 100%: Ion suppression

Acceptance Criteria (typical)

According to ICH M10 and FDA guidance:

  • ME should be between 80-120% (±20%)

  • CV of ME should be ≤15%

Experimental Design

To assess matrix effects:

  1. Prepare blank matrix (e.g., plasma) from multiple sources

  2. Spike analyte post-extraction at known concentration

  3. Compare to analyte in neat solvent at same concentration

See also

step_measure_standard_addition(), measure_accuracy()

Other calibration: step_measure_dilution_correct(), step_measure_standard_addition(), step_measure_surrogate_recovery()

Examples

# Matrix effect study data
me_data <- data.frame(
  sample_type = rep(c("matrix", "neat"), each = 6),
  matrix_lot = rep(c("Lot1", "Lot2", "Lot3", "Lot1", "Lot2", "Lot3"), 2),
  concentration = rep(c("low", "high"), each = 3, times = 2),
  response = c(
    # Matrix samples (some suppression)
    9500, 9800, 9200, 48000, 49500, 47000,
    # Neat samples
    10000, 10000, 10000, 50000, 50000, 50000
  )
)

me <- measure_matrix_effect(
  me_data,
  response_col = "response",
  sample_type_col = "sample_type",
  matrix_level = "matrix",
  neat_level = "neat",
  concentration_col = "concentration"
)

print(me)
#> measure_matrix_effect
#> ──────────────────────────────────────────────────────────────────────────────── 
#> 
#> Overall Matrix Effect Summary:
#>   Groups evaluated: 2 
#>   Mean ME: 96 %
#>   SD ME: 0.9 %
#>   CV ME: 1 %
#>   Range: 95 - 96 %
#> 
#> Effect Classification:
#>   Ion suppression (ME < 100%): 2 
#>   Ion enhancement (ME > 100%): 0 
#>   Acceptable (80-120%): 2 / 2 
tidy(me)
#>   concentration n_matrix n_neat mean_matrix_response mean_neat_response
#> 1           low        3      3              9500.00              10000
#> 2          high        3      3             48166.67              50000
#>   matrix_effect_pct me_ci_lower me_ci_upper interpretation
#> 1          95.00000    87.54759    102.4524    suppression
#> 2          96.33333    90.08172    102.5849    suppression