rumenGP: An R package for processing, modeling, and comparing in vitro rumen gas production data

In vitro gas production is widely used to characterize rumen fermentation kinetics, but data processing and model selection often rely on spreadsheets and custom scripts with limited reproducibility. We developed rumenGP, an open-source R package that integrates data import, validation, quality control, kinetic modeling, model comparison, visualization, and reporting for in vitro gas-production data. The package supports pressure-based (ANKOM RF) and generic workflows, implements 15 nonlinear models, and allows custom model fitting. Model performance is evaluated using RSS, R², RMSE, AIC, and BIC. An example dataset containing 25 bottles across five treatment groups was used to demonstrate package functionality. After quality-control procedures, 21 bottles and 3,038 observations were retained, and all 15 models converged successfully. Diagnostic evaluation highlighted differences among models in goodness of fit and parameter behavior, emphasizing the importance of combining convergence assessment with model diagnostics. rumenGP provides a transparent and reproducible framework for analyzing in vitro rumen gas-production experiments and comparing alternative kinetic models.