OrganScaleR: An Open Shiny Tool for Principled Organ-Weight Inference in Mouse Physiology
BackgroundOrgan weights are often divided by body weight to report "relative" organ size. Yet ratios usually stay size-dependent and become misleading. We built a simple decision path for size adjustment and wrapped it in a Shiny application so physiologists can get correct answers without coding. MethodsWe reused liver and body-weight data from a mouse nutrition study for confirmatory examples in two common cases: when diet groups shared a wide body-size range, and when diet produced much smaller animals with little overlap. We compared liver-to-body-weight ratios with size-adjusted regression when the body-size overlap allowed comparisons, and with causal Bayesian mediation when it did not. The full workflow was implemented in OrganScaleR, a guided R Shiny application. ResultsThe ratio-normalized organ weight is still associated with body weight, leading to misleading comparisons. Modeling liver weight against body weight gave more cautious, size-adjusted effects when groups shared a common size range. When diets shifted body size strongly and overlap was limited, causal mediation showed that most organ differences followed the change in body weight rather than an organ-specific action. Simulations confirmed that ratios can generate false positives and biased estimates under allometric scaling, while model-based approaches remained reliable. OrganScaleR implements this decision workflow in a guided Shiny application that returns interpretable effects. ConclusionsOrganScaleR selects scale, enforces common support, and routes the analysis to size-adjusted ANCOVA or causal Bayesian mediation depending on body-weight overlap. It reports adjusted effects in original units through a point-and-click workflow, removing the statistical barrier to abandoning ratio normalization. HighlightsO_LIBody-weight ratios remain size-dependent and distort organ-weight comparisons. C_LIO_LIANCOVA at a common reference body weight removes ratio bias when groups overlap. C_LIO_LICausal mediation separates organ-specific from body-weight-mediated diet effects. C_LIO_LISimulations confirm ratios inflate false-positive rates under allometric scaling. C_LIO_LIOrganScaleR guides size adjustment without coding via a point-and-click workflow. C_LI