Statistical modeling of gut microbiota for personalized health status monitoring

Microbiome

Table 1 Health prediction results using different methods

Method	Feature number	Discovery dataset			Validation dataset
Method	Feature number	Healthy	Unhealthy	Average	Healthy	Unhealthy	Average
RF	313	1.0000	1.0000	1.0000	0.8220	0.5350	0.6780
XGB	313	1.0000	1.0000	1.0000	0.7120	0.5800	0.6460
GMHI	50	0.7560	0.6376	0.6970	0.7712	0.6220	0.6966
hiPCA	313	0.6988	0.7680	0.7334	0.5678	0.7259	0.6469
RF-hiPCA	50	0.7117	0.6645	0.6881	0.3305	0.7605	0.5455
XGB-hiPCA	50	0.7595	0.6078	0.6837	0.7119	0.5422	0.6270
GMHI-hiPCA	50	0.7242	0.6885	0.7063	0.7034	0.6461	0.6747
MIC-hiPCA	50	0.7303	0.6511	0.6907	0.3305	0.7244	0.5275
SPM-hiPCA	50	0.7470	0.6727	0.7098	0.6780	0.6235	0.6507
KS-hiPCA	50	0.7128	0.7031	0.7080	0.6864	0.6732	0.6798
KS-hiPCA	92	0.7140	0.7124	0.7132	0.7034	0.7018	0.7026
MRS-GMHI	6	0.6495	0.6172	0.6334	0.5508	0.6054	0.5781
MRS-hiPCA	6	0.8524	0.3641	0.6083	0.8475	0.3283	0.5879

ISSN: 2049-2618