The study uses machine learning methods to examine the effects of genotype and behavioral interventions on the patterns of protein expression in mice. The dataset includes measurements of the expressions levels of 77 proteins in the cerebral cortex of 72 mice, comprising 34 trisomic (Down syndrome) mice and 38 normal mice. A total of 1080 measurements were made for each protein, with 15 measurement per sample. Based on genotype, behavior, and treatment, the mice are divided into eight different groups: trisomy mice, control mice stimulated to learn and injected with saline (c-CS-s), control mice stimulated to learn and injected with memantine (c-CS-m), and control mice not stimulated to learn and injected with saline (c-SC-s). Memantine was injected into control mice that had not been stimulated to learn (c-SC-m), memantine was injected into trisomy mice that had been stimulated to learn and given saline (t-CS-s), memantine was injected into trisomy mice that had been stimulated to learn and given memantine (t-CS-m), and trisomy mice had not been stimulated to learn and given saline (t-SC-s). In this work, four distinct classification algorithms-Decision Tree, Random Forest, Support Vector Machine, and Gradient Boosting-were used to sort the mice into their appropriate classifications. The goal was to evaluate each algorithm’s capability and accuracy in assigning precise class labels using protein expression data. We assessed the effectiveness and accuracy of different algorithms in completing this categorization assignment through a comparative study.