Decision tree-based method for integrating gene expression, demographic, and clinical data to determine disease endotypes

Table 3 Modk-prototypes weighting schemes

Gene expression	Clinical covariates	Indicators of disease status
33	33	33
20	40	40
40	20	40
50	50	0
30	60	10
60	30	10
40	40	20
Adaptive	Adaptive	Adaptive

The Modk algorithm was run with 7 pre-defined weighting schemes. This table shows the combinations of gene expression, clinical covariates, and disease indicator weighting for each run. For the adaptive weighting scheme, the weights were determined by the algorithm as described in the methods.