FIGURE 1.

Entire procedures of the investigation on DNA methylation profile of patients with Alzheimer’s disease and control samples. The profile is retrieved from Gene Expression Omnibus. Four datasets are constructed, corresponding to four brain structures. Each dataset is deeply analyzed by Boruta feature filtering and max-relevance and min-redundancy methods one by one, resulting in a feature list. This list is used in the incremental feature selection, containing some classification algorithms, to identify methylation signatures, rules and construct efficient classifiers.