Fig. 1

A Overview of CimpleG and statistics used for feature selection and downstream applications. B–C Principal component analysis of the DNAm datasets with somatic cells (B) or leukocytes (C). Only cell types highlighted in bold, for which we have samples in train (\(\ge 10\) positive examples) and test data, were used as target classes. Cell types that are not present in the test data are only used as negative examples (non-target cells)