Getting Started with LactCurveModels

Overview

LactCurveModels is an R package for fitting nonlinear lactation curve models to daily milk yield records from dairy animals. The package fits up to 20 published models simultaneously, evaluates each model using six goodness-of-fit metrics, and produces publication-ready diagnostic figures and summary tables — all in a single function call.

Key features:

• Fits 20 nonlinear lactation curve models spanning four mathematical families
• Supports batch processing of multiple animals from a single CSV file
• Produces 15 diagnostic figures per animal plus 3 combined cross-animal outputs
• Reports R², Adjusted R², RMSE, AIC, BIC, and Durbin-Watson for every model
• Flexible selection of any subset of models or animals to analyse

Installation

Install the package from CRAN:

install.packages("LactCurveModels")

Or install the development version from GitHub:

# install.packages("devtools")
devtools::install_github("venkatesanraja/LactCurveModels")

Load the package:

library(LactCurveModels)

Input Data Format

Your input CSV file must contain exactly three columns in any column order:

A typical CSV file looks like this:

animal_id,dim,dmy
Animal_A,1,9.3
Animal_A,2,9.9
Animal_A,3,10.2
...
Animal_B,1,8.1
Animal_B,2,8.7
...

Note: The dim column represents a fortnightly time index across the lactation period. Each record corresponds to one fortnightly observation of daily milk yield for a given animal.

Basic Usage

results <- run_lactation_analysis(
  input_csv_path = "path/to/your/data.csv"
)

results <- run_lactation_analysis(
  input_csv_path = "path/to/your/data.csv",
  out_dir        = "D:/my_results/lactation_analysis"
)

Selecting Specific Models

Use the selected_models argument to fit only a subset of the 20 available models. This is useful when you already know which model families are most appropriate for your data, or when you want faster processing.

results <- run_lactation_analysis(
  input_csv_path  = "path/to/your/data.csv",
  selected_models = c("Wood_1967", "Wilmink_k005", "Ali_Schaeffer",
                      "Pollott_Multiplicative")
)

Selecting Specific Animals

Use the selected_animals argument to analyse only a subset of the animals present in the CSV file. Animal IDs must match the values in the animal_id column exactly.

results <- run_lactation_analysis(
  input_csv_path   = "path/to/your/data.csv",
  selected_animals = c("Animal_A", "Animal_C")
)

Combined Selection of Models and Animals

All four arguments can be used together:

results <- run_lactation_analysis(
  input_csv_path   = "path/to/your/data.csv",
  selected_models  = c("Wood_1967", "Wilmink_k005", "Ali_Schaeffer"),
  selected_animals = c("Animal_A", "Animal_B"),
  out_dir          = "D:/results"
)

Understanding the Outputs

Accessing Results in R

The function returns a named list (one element per animal) which can be accessed directly in R:

# View goodness-of-fit metrics for Animal_A
results[["Animal_A"]]$metrics_df

# View parameter estimates for Animal_A
results[["Animal_A"]]$param_table

# Access the fitted nls object for Wood (1967) model
results[["Animal_A"]]$model_fits[["Wood_1967"]]$model

# View model coefficients
coef(results[["Animal_A"]]$model_fits[["Wood_1967"]]$model)

# View individual metrics
results[["Animal_A"]]$model_fits[["Wood_1967"]]$metrics

Using fit_lactation_models() Directly

For advanced users, the lower-level function fit_lactation_models() fits models to a single animal’s pre-formatted data frame:

# Build a data frame for one animal
animal_df <- data.frame(
  x = 1:20,
  y = c(9.3, 9.9, 10.2, 10.4, 10.3, 10.1, 9.9, 9.6, 9.3, 9.0,
        8.7, 8.4, 8.1, 7.8, 7.5, 7.2, 6.9, 6.6, 6.3, 6.0),
  z = (1:20) / 365
)

# Fit all 20 models
fits <- fit_lactation_models(animal_df)

# Fit selected models only
fits <- fit_lactation_models(
  animal_df,
  selected_models = c("Wood_1967", "Wilmink_k005", "Ali_Schaeffer")
)

# Access results
fits[["Wood_1967"]]$metrics
fits[["Wood_1967"]]$predictions
coef(fits[["Wood_1967"]]$model)

The data frame passed to fit_lactation_models() must contain:

Goodness-of-Fit Metrics

The following metrics are computed for every successfully fitted model:

Citation

If you use LactCurveModels in your research, please cite it as:

citation("LactCurveModels")

Raja TV, Lalremruati PC, Priyadharshini P, Nidhishree NS, Gurjar D, Alex R, Vohra V (2025). LactCurveModels: Lactation Curve Model Fitting for Dairy Animals. R package version 0.1.0. https://cran.r-project.org/package=LactCurveModels

Getting Help

• Use ?run_lactation_analysis and ?fit_lactation_models for function-level help
• Report bugs at: https://github.com/venkatesanraja/LactCurveModels/issues