MeatColor is an R package for turning instrumental meat-color measurements into analysis-ready summaries, interpretable comparisons, and publication-ready graphics. It works with CIE L*a*b* colorimeter data and spectral reflectance measurements, bringing common meat-science calculations into one reproducible workflow.
Instrumental color data are easy to collect but often tedious to analyze. Researchers must summarize repeated measurements, convert coordinates into meaningful color metrics, compare treatments, apply specialized myoglobin equations, and build figures—usually with separate formulas and scripts. MeatColor connects those steps while keeping the data in ordinary R data frames and returning plots that can be customized with ggplot2.
- Built for meat-color experiments. Group measurements by treatment, product, storage time, or other experimental factors.
- More than L*a*b* averages. Calculate hue, chroma, CIEDE2000 color differences, and surface myoglobin redox-form estimates.
- Treatment-aware comparisons. Compare every sample pair, summarize within- and between-treatment distances, and visualize the results.
- Transparent scientific behavior. Input scales and out-of-range handling are explicit; questionable estimates are not silently hidden.
- Fits existing R workflows. Functions accept data frames, support tidy column selection where appropriate, and return regular data frames or ggplot objects.
| Research task | Main functions | Result |
|---|---|---|
| Summarize and display instrumental color | summarize_lab(), plot_lab_colors() |
Treatment-level L*a*b* summaries and color charts |
| Calculate color attributes | lab_chroma(), lab_hue(), add_lab_metrics() |
Chroma and hue-angle variables |
| Approximate measured color on screen | lab_to_hex() |
CIELAB-to-sRGB hexadecimal approximations |
| Quantify perceptual color differences | delta_e_2000(), lab_distances() |
Paired or all-pairs CIEDE2000 distances |
| Compare treatments | summarize_treatment_distances(), plot_treatment_distances() |
Treatment summaries and heatmaps |
| Estimate myoglobin redox forms | myoglobin_int(), myoglobin_ref() |
OMb, DMb, and MMb estimates from reflectance |
| Reuse prepared-reference calibrations | myoglobin_calibration(), predict() |
Validated calibration objects and predictions |
MeatColor requires R 4.1.0 or later. Install the development version from GitHub with:
# install.packages("pak")
pak::pak("JustSplash8501/MeatColor")Start with a data frame containing L*, a*, and b* measurements plus the variables that define the experiment:
library(MeatColor)
measurements <- data.frame(
sample = paste0("S", 1:8),
treatment = rep(c("Control", "Aged"), each = 4),
day = rep(c("Day 0", "Day 7"), each = 2, times = 2),
l = c(45.2, 44.8, 42.1, 41.9, 46.1, 45.9, 43.2, 42.8),
a = c(18.5, 18.8, 16.2, 16.5, 17.9, 18.1, 15.8, 16.1),
b = c(12.1, 12.3, 10.5, 10.7, 11.8, 12.0, 10.3, 10.5)
)
color_summary <- summarize_lab(
measurements,
group_vars = c("treatment", "day")
)
plot_lab_colors(
color_summary,
x_var = "day",
group_var = "treatment",
x_label = "Storage time",
group_label = "Treatment",
title = "Instrumental meat color over time",
show_values = "lab"
)summarize_lab() calculates group means and display-color approximations.
plot_lab_colors() turns those results into a treatment-aware chart and
returns a regular ggplot object, so themes, labels, scales, and other ggplot2
layers can be added normally.
This figure shows instrumental color across wet-aging durations in a published beef study. It illustrates how MeatColor can make changes among treatments and time points visually comparable while retaining the underlying quantitative measurements.
Source: Main, A. J., Frink, L. M., Hernandez, M. S., O'Quinn, T. G., Legako, J. F., Miller, R. K., Nair, M. N., Kerth, C. R., Lancaster, J. M., & Woerner, D. R. (2026). “Extended Beef Wet-Aging Influences on Biceps femoris, Gluteus medius and Semimembranosus Palatability.” Meat and Muscle Biology, 10(1), 22594, 1–19. https://doi.org/10.22175/mmb.22594
CIELAB coordinates can look different without revealing how large that
difference is perceptually. lab_distances() calculates CIEDE2000 (Delta E 00)
for every sample pair, without requiring a designated reference sample:
distances <- lab_distances(measurements, sample_id = sample)
# Symmetric sample-by-sample matrix
as.matrix(distances)
# Unique within- and between-treatment comparisons
treatment_distances <- summarize_treatment_distances(
distances,
treatment = treatment
)
# Treatment-level CIEDE2000 heatmap
plot(distances, treatment = treatment)Self-comparisons and duplicate A–B/B–A combinations are excluded from
treatment summaries. The lower-level delta_e_2000() function remains
available when colors are deliberately paired or a true reference color
exists.
MeatColor supports two reflectance-based approaches for estimating relative oxymyoglobin (OMb), deoxymyoglobin (DMb), and metmyoglobin (MMb):
myoglobin_int()applies the selected-wavelength reflex-attenuance equations to MiniScan reflectance data.myoglobin_ref()applies calibrated K/S equations using experimentally prepared 100% OMb, DMb, and MMb reference spectra.
For studies that repeatedly use the same prepared references,
myoglobin_calibration() creates a validated object that can be applied to
multiple sample data sets with predict():
calibration <- myoglobin_calibration(
omb_reference = omb_reference_scans,
dmb_reference = dmb_reference_scans,
mmb_reference = mmb_reference_scans,
reflectance_scale = "percent"
)
myoglobin_results <- predict(calibration, newdata = sample_scans)
plot(calibration)Reference scans and study samples should be collected with the same product, instrument settings, standardization, and experimental conditions. Estimates outside 0–100% are retained with a warning by default rather than silently altered.
MeatColor implements established color-science methods while making important assumptions visible to the analyst:
- CIELAB display colors are approximations. Physical L*a*b* measurements are converted to sRGB for visualization; some measured colors fall outside the displayable sRGB gamut.
- Color differences use CIEDE2000. The implementation follows Sharma, Wu, and Dalal (2005), with configurable lightness, chroma, and hue weighting factors. https://doi.org/10.1002/col.20070
- Myoglobin calculations follow AMSA guidance. The package implements both selected-wavelength and prepared-reference approaches described in the AMSA Meat Color Measurement Guidelines.
See the conversion formulation for the CIELAB-to-CIEXYZ-to- sRGB mathematics. Researchers remain responsible for choosing a method and instrument configuration appropriate to their product and study design.
References: American Meat Science Association. (2012). Meat Color Measurement Guidelines (revised December 2012). View the guidelines. King, D. A., et al. (2023). “American Meat Science Association Guidelines for Meat Color Measurement.” Meat and Muscle Biology, 6(4), 1–81. https://doi.org/10.22175/mmb.12473
The hosted LAB Color Explorer helps students connect L*, a*, and b* coordinates with an approximate screen color. Sliders and numeric inputs make each axis immediately visible:
- L* represents lightness, from 0 (black) to 100 (white).
- a* moves from green at negative values to red at positive values.
- b* moves from blue at negative values to yellow at positive values.
The explorer also identifies out-of-gamut colors, demonstrating why some physical CIELAB measurements cannot be represented exactly on a screen. The teaching app is hosted separately; its source code is not part of this package.
- Start with the introductory vignette for a fuller analysis workflow.
- Use
help(package = "MeatColor")or?function_namefor function-level documentation. - Report reproducible problems through the issue tracker.
- Read CONTRIBUTING.md before proposing substantial changes.
MeatColor 0.2.0 is the first formal release. Its documented interfaces are intended for regular research use and will be evolved through documented, versioned releases. Scientific methods should still be selected and interpreted in the context of the instrument, product, and study design.
MeatColor is independently maintained and is not affiliated with or endorsed by any institution or company.
Participation in this project is governed by the Contributor Covenant Code of Conduct.