The following is an example of a clinical study aimed at identification of circulating metabolites related to disease phenotype or grade/severity/type (tissue histology, 4 classifications including controls).
The challenge is to make sense of 300 metabolic measurements for 300 patients.
The goal is to identify metabolites related to disease, while accounting covariate meta data such as gender and smoking.
- Exploratory Data Analysis – principal components analysis (PCA)
- Statistical Analysis – covariate adjustment and analysis of covariance or ANCOVA
- Multivariate Classification Modeling – orthogonal signal correction partial least squares discriminant analysis (O-PLS-DA)
Data exploration is useful for getting an idea of the data structure and to identify unusual or unexpected trends.
PCA above conducted on autoscaled data (300 samples and 300 measurements) was useful for identifying an interesting 2-cluster structure in the sample scores (top left). Unfortunately the goal of the study, disease severity, could not explain this pattern (top center). An unknown covariate was identified causing the observed clustering of samples (top right).
Next various covariate adjustment strategies were applied to the data and evaluated using the unsupervised PCA (bottom left) and the supervised O-PLS-DA.
Even after the initial covariate adjustment for the 2-cluster effect there remained a newly visible covariate (top ;left), the source of which could not me linked to the meta data.
After data pre-treatment and evaluation of testing strategies (top right) the next challenge is to select the best classifiers of disease status. Feature selection was undertaken to improve model performance and simplify its performance.
Variable correlation with O-PLS-DA sample scores and magnitude of variable loading in the model were used to select from the the full feature set (~300) only 64 (21%) top features which explained most of the models classification performance.
In conclusion preliminary data exploration was used to identify an unknown source of variance which negatively affected the experimental goal to identify metabolic predictors of disease severity. Multivariate techniques, PCA and O-PLS-DA, were used to identify an optimal data covariate adjustment and hypothesis testing strategy. Finally O-PLS-DA modeling including feature selection, training/testing validations (n=100) and permutation testing (n=100) were used to identify the top features (21%) which were most predictive of patients classifications as displaying or not displaying the disease phenotype.
A typical experiment may involve the testing of a wide variety of factors. For instance, here is an example of an experiment aimed at determining metabolic differences between two plant cultivars at three different ontological stages and in two different tissues. Exploratory principal components analysis (PCA) can be used to evaluate the major modes of variance in the data prior to conducting any univariate tests.
Based on the PCA (autoscaled data) we can see that the majority of the differences are driven by differences between tissues. This is evident from the scores separation in (a) between leaf and fruit tissues, which is driven by metabolites with large positive/negative loadings on the first dimension or x-axis in (b). A lesser mode of variance is captured in the second dimension, and particularly in fruit we can see that there is some separation in scores between the two cultivars and their different ontological stages. Based on this it was concluded to carry out test in leaf and fruit tissue separately. Additionally in order to identify the effects of cultivar on the metabolomic profiles which are independent of stage and vice versa, a linear covariate adjustments were applied to the data.
Again using PCA and focusing on fruit tissue, we can evaluate the variance in the data given our hypotheses (differences between cultivars or stages). Looking at (a) we can see that there is not a clear separation in scores in any one dimension between cultivars or stages. However there is separation in two dimensions. This is problematic in that this suggest that there is an interaction between cultivar and stage, which will complicate any univariate tests for these factors. We can see that carrying out linear covariate adjustment either for cultivar (b) or stage (c) translate the variance for the target hypothesis into one dimension, which therefore simplifies its testing. Note, this is exactly what is done when doing an analysis of covariance or ANCOVA. However if we want to use this same favorable variance environment for multivariate modeling like for example partial least squares projection to latent structures discriminant analysis (PLS-DA) we need to covariate adjust the data which in this case is achieved by taking the residuals from linear model for the covariate we want to adjust for.
Now that we have adjusted the data for covariate effects we can test the primary hypotheses (differences between cultivars, stages and tissues) using PLS-DA. Quick visual inspection of model scores can be used to get a feel for the quality of the models. Ideally we would like to see a scores separation between the various levels of our hypotheses in one dimension. We can see that both fruit models are of higher quality than that for leaf. However to fully validate these models we need to carry out some permutation testing or something similar. The benefit of PLS-DA is that we can use the information about the variables contribution to the scores separation or loadings to identify metabolomic differences between cultivars or with increasing maturity or stage.
Here is an example where PLS-DA variable loadings are mapped into a biochemical context using a chemical similarity network. This network represents differences in metabolites due to cultivar, wherein significant differences in metabolite means (ANCOVA, FDR adjusted p-value < 0.05) between cultivars are represented by nodes or vertices which are colored based on the sign of the loading and their size used to encode the magnitude of their loading in the model.
We can now compare the two networks representing metabolomic differences due to cultivar (far top) or to stage (above) to identify biochemical changes due to these factors which are independent of each others effects (or interaction).