This article introduces a new method for detecting outliers in nonlinear regression analysis, called the ROUT method. It combines robust nonlinear regression with the false discovery rate (FDR) approach to identify outliers. The method first fits data using a robust nonlinear regression assuming a Lorentzian distribution of residuals. It then uses the FDR approach to identify outliers, which are defined as points that are too far from the curve. Outliers are removed, and the data are re-analyzed using ordinary least-squares regression. The ROUT method is designed to identify outliers with reasonable power and few false positives. The method is tested on simulated data and shown to detect outliers with an average false discovery rate of less than 1%. It is also effective at identifying outliers in data contaminated with one or more outliers. The method is robust to outliers and is not affected by the choice of initial parameter values. It is also adaptive, meaning it becomes more robust as the fit improves. The method is implemented in GraphPad Prism version 5 and is available for use in nonlinear regression analysis. The method is recommended for use when analyzing data with a Gaussian distribution and a small number of outliers. It is also useful for detecting outliers in data with non-Gaussian scatter. The method is not recommended for use when the data are not independent or when the number of outliers is large. The method is also not recommended for use when the data are not well fitted by the model. The method is designed to be used when the model is correct and the data are independent. The method is not recommended for use when the data are not independent or when the number of outliers is large. The method is also not recommended for use when the data are not well fitted by the model. The method is designed to be used when the model is correct and the data are independent.This article introduces a new method for detecting outliers in nonlinear regression analysis, called the ROUT method. It combines robust nonlinear regression with the false discovery rate (FDR) approach to identify outliers. The method first fits data using a robust nonlinear regression assuming a Lorentzian distribution of residuals. It then uses the FDR approach to identify outliers, which are defined as points that are too far from the curve. Outliers are removed, and the data are re-analyzed using ordinary least-squares regression. The ROUT method is designed to identify outliers with reasonable power and few false positives. The method is tested on simulated data and shown to detect outliers with an average false discovery rate of less than 1%. It is also effective at identifying outliers in data contaminated with one or more outliers. The method is robust to outliers and is not affected by the choice of initial parameter values. It is also adaptive, meaning it becomes more robust as the fit improves. The method is implemented in GraphPad Prism version 5 and is available for use in nonlinear regression analysis. The method is recommended for use when analyzing data with a Gaussian distribution and a small number of outliers. It is also useful for detecting outliers in data with non-Gaussian scatter. The method is not recommended for use when the data are not independent or when the number of outliers is large. The method is also not recommended for use when the data are not well fitted by the model. The method is designed to be used when the model is correct and the data are independent. The method is not recommended for use when the data are not independent or when the number of outliers is large. The method is also not recommended for use when the data are not well fitted by the model. The method is designed to be used when the model is correct and the data are independent.