Use of SNP genotypes to identify carriers of harmful recessive mutations in cattle populations

BMC Genomics

Table 4 Variance and bias of prediction error

	Method	TER			FNR			FPR
		Error	Var	Bias	Error	Var	Bias	Error	Var	Bias
Brown	KNN	16.3 %	0.03306	0.13923	13.8 %	0.02989	0.09918	20.6 %	0.02136	0.21079
	Lasso	1.02 %	0.00031	0.00962	0.00 %	-	-	3.02 %	0.00087	0.02653
	SVML	0.91 %	0.00161	0.00873	0.12 %	0.00101	0.00021	2.41 %	0.00267	0.02367
	SVMR	2.15 %	0.00221	0.20115	1.11 %	0.00258	0.02264	3.92 %	0.00153	0.15731
	MAG	1.71 %	0.00423	0.01513	0.81 %	0.00010	0.01180	1.62 %	0.01178	0.02107
Fleckvieh	KNN	2.96 %	0.00143	0.02981	71.8 %	0.02688	0.72386	0.16 %	0.00021	0.00167
	Lasso	0.22 %	0.00011	0.00192	4.48 %	0.00115	0.04245	0.04 %	0.00007	0.00015
	SVML	0.37 %	0.00043	0.00251	7.63 %	0.04131	0.04083	0.07 %	0.00014	0.00093
	SVMR	0.93 %	0.00046	0.00571	22.5 %	0.01024	0.09911	0.05 %	0.00011	0.00186
	MAG	0.64 %	0.00094	0.00487	0.98 %	0.00101	0.06040	0.24 %	0.00097	0.00252

KNN k-nearest neighbors, Lasso ℓ1-penalized logistic regression, SVML support vector machine with linear kernel, SVMR support vector machine with radial kernel, MAG multi-allelic gene prediction

ISSN: 1471-2164