A novel meta-analysis based on data augmentation and elastic data shared lasso regularization for gene expression

BMC Bioinformatics

Table 7 Discrimination results from all methods

Method	Training error	Validation accuracy (%)	No. of selected genes	Convergence time (s)
Lasso	2.41% (0.001)	94.03	33	2.99
Elastic Net	2.10% (0.002)	95.48	65	5.67
HLR	1.88% (0.002)	93.31	52	63.80
SGL	1.62% (0.001)	96.70	67	56.71
Meta-Lasso	2.13% (0.001)	93.63	50	72.05
DSL	2.01% (0.000)	95.82	46	17.50
DA-DSL-L₂	1.15% (0.000)	98.39	51	43.88

ISSN: 1471-2105