Fig. 5
(A) PCA and (B) PLS-DA scores plots indicating the distribution of all samples (healthy controls and all untreated and treated patient samples) after sample exclusions (n = 65). Post sample exclusion, the groups were more homogenous and could be better distinguished from each other, more so with PLS-DA analysis and those variables important in separating the groups, identified. The PLS-DA model (B) did not validate (cross-validation for a one-component model: Q2 = 0.40, R2 = 0.60, accuracy = 0.60; permutation testing p-value = 0.241) and was used solely in an explorative, as opposed to predictive manner