Knowledge transfer via classification rules using functional mapping for integrative modeling of gene expression data

Table 10 Comparing average performance per disease type to merged datasets per disease type. This table shows the average classification performance per disease type as compared to merged datasets per disease type. In the dataset column, Avg denotes average, MM denotes merged by meta-analysis, and M means merged by cross-platform data merging

Dataset	SVM	LDA	RF	C4.5	NB	PLR	RL	TRL	TRL-FM
Average performance per disease type
Avg_brain	0.67	0.66	0.66	0.64	0.73	0.69	0.66	0.68	0.76
Avg_ipf	0.80	0.78	0.78	0.78	0.82	0.86	0.85	0.89	0.95
Avg_prostate	0.89	0.83	0.83	0.76	0.88	0.90	0.82	0.86	0.89
Merged per disease type by meta-analysis
MM_brain	0.67	0.70	0.70	0.69	0.70	0.69	0.67	*	*
MM_ipf	0.88	0.88	0.88	0.85	0.74	0.88	0.81	*	*
MM_prostate	0.89	0.84	0.84	0.81	0.70	0.85	0.76	*	*
Merged per disease type by batch effect removal
M_Brain	0.50	0.51	0.51	0.48	0.53	0.51	0.54	*	*
M_IPF	0.67	0.63	0.63	0.60	0.63	0.64	0.68	*	*
M_Prostate	0.53	0.53	0.53	0.53	0.53	0.55	0.59	*	*

*denotes that transfer learning methods were not evaluated. Currently, TRL and TRL-FM cannot be applied to cross-domain studies (i.e., transfer from one disease type to another)

ISSN: 1471-2105