Best fit results using multifactor dimensionality reduction analysis of one-, two-, and three-locus models for CAL formation in KD patients.

a Average testing balanced accuracy is the accuracy of classifications for cases and controls in the testing dataset (one-twentieth of the data) calculated as (Sensitivity+Specificity)/2.
b Average cross validation consistency is the number of times the model was selected as the best model after 20 cross-validation runs.
c Significance of accuracy (empirical p-value based on 10,000 permutations).
** p-value ≤ 0.05
Best fit results using multifactor dimensionality reduction analysis of one-, two-, and three-locus models for CAL formation in KD patients.