In Statistics, it is often of interest to compare means from two or more populations. This is usually carried out by the method called Analysis of Variance (ANOVA). However, the comparison becomes complicated when the data to analyze is assumed to have a functional form. In this talk, we introduce a test analogous to the classical ANOVA which is based on a distribution-free procedure to compare similarity between sets of curves. We employ the constructed test statistic on a study comparing ganglioside profiles (complex brain proteins) in rats. We provide the asymptotic distribution of the proposed test statistic and validate these results with a small simulation study.