Additional file 1: Figure S1. Pathway of the linear open system. Figure S2. Time profiles for the data generated by the linear open system with a specific setting of parameters and the factor plot of the Paralind(1,2,2) model for this dataset. Figure S3. Pathway of the glycolysis model. Figure S4. Cross-validation performance of the 2-component CP model and the Paralind(1,2,2) model for data generated by the glycolysis model with one source of induced variation and the level of the individual variation as 0.02. Figure S5. Comparison of the true and predicted data by Paralind(1,2,2) and Paralind(1,3,3) for metabolite G6P. Figure S6. Time profiles for the data generated by the glycolysis model with one source of induced variation and the level of the individual variation as 0.5, as well as the factor plot of the 2-component CP model for this dataset. Figure S7. Time profiles for the data generated by the glycolysis model with 100 subjects and the level of the individual variation as 0.5, as well as the factor plot of the 3-component CP model for this dataset. Figure S8. Pathway of the cholesterol model. Figure S9. Cross-validation performance of the 2-component CP model and the Paralind(1,2,2) model for data generated by the cholesterol model with one source of induced variation and the level of the individual variation as 0.02. Table S1. Model selection information, including fit, CC, TC, C12 values, for CP models applied to the data generated by the glycolysis model with one source of induced variation and the level of the individual variation as 0.36. Table S2. Model selection information, including fit, CC, TC, C12 values, for CP models applied to the data generated by the cholesterol model with one source of induced variation and the level of the individual variation as 0.02. Table S3. Model selection information, including fit, CC, TC, C12 values, for CP models applied to the data generated by the cholesterol model with one source of induced variation and the level of the individual variation as 0.65. Table S4. Model selection information, including fit, CC, TC, C12 values, for CP models applied to the data generated by the cholesterol model with two sources of induced variations and the level of the individual variation as 0.02.