Fig 1.
A flow chart to describe the classification process.
Fig 2.
Batch effects and their COMBAT adjustment on merging Nelson and Rothman datasets.
Table 1.
The SR of the optimised feature lists on the NelsonB and Rothman datasets.
Table 2.
The classification statistics, including SR, sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) were recorded after the classification of each dataset.
Table 3.
Upper limit of 95% confidence interval (CI Upper) and lower limit of 95% confidence interval (CI Lower) for SR, sensitivity, specificity, PPV and NPV were recorded after the classification of each dataset.
Fig 3.
The ROC graph is plotted to show the performance of the binary classifiers.
Fig 4.
A MDS plot created for Beata dataset to show the distribution of its cases and controls.