When evaluating LR and RF on this dataset using the same evaluation procedure as for the OpenML datasets, we see that LR and RF perform very similarly for all three considered measures: 0.722 for LR versus 0.729 for RF for the accuracy (acc), 0.792 for LR versus 0.785 for RF for the Area Under the Curve (auc) and 0.185 for LR versus 0.187 for RF for the Brier score. When looking at permutation variable importances (for RF) and p-values of the Wald test (for LR), we see that the 13 candidate features are assessed similarly by both methods. In particular, the two closest neighbor nucleotides are by far the strongest predictors for both methods.
