Figure 11.

Analysis of models with varying numbers of regressors. Shown are the results of best subset regression algorithm which evaluates the best models of 1, 2, . . ., n regressors. (A) The residual sum of squares of the best models of A549 data as a function of the number of components (regressors). There is no significant decrease in the residual sum of squares for models with more than 10 regressors. (B) The residual sum of squares of the best models of AG02603 data as a function of the number of components (regressors). There is no significant decrease in the residual sum of squares for models with more than 7 regressors. (C) The regression coefficients for the best 10 component model of A549 data. The components involve single drug concentrations, pairwise and three-drug interactions. (D) The regression coefficients for the best 7 component model of AG02603 data. The components involve single drug concentrations and pairwise interactions.

Al-Shyoukh et al. BMC Systems Biology 2011 5:88   doi:10.1186/1752-0509-5-88
Download authors' original image