Skip to main content

Table 6 Descriptive statistics for predictive accuracy (estimates less than 0 were set to 0 whereas estimates greater than 1 were set to 1) by scenario

From: Evaluation of approaches for estimating the accuracy of genomic prediction in plant breeding

  

Methods

Scenario

Statistic

M0

*M1(15)

M2(16)

M3(17)

M4(21)

M5(24)

M6(25)

M7(35)

  

r g , g ^

r g ^ , p H ^ m 1

r g ^ , p H ^ m 2

r g ^ , p H ^ m 3

r g ^ , p H ^ m 4

H ^ m 5

r g , g ^ , m 6

ρ ^ m 7

1

N

1000

1000

1000

1000

1000

1000

1000

1000

 

MIN

0.750

0.327

0.265

0.265

0.384

0.707

0.316

0.750

 

MEAN

0.843c

0.877a

0.727 e

0.727e

0.858b

0.819d

0.663f

0.840c

 

MAX

0.908

1.000

1.000

1.000

1.000

0.893

0.884

0.899

 

STD

0.024

0.109

0.104

0.104

0.093

0.028

0.084

0.023

 

MSD

0.000

0.013

0.025

0.025

0.009

0.002

0.040

0.001

 

Q1

0.829

0.807

0.661

0.661

0.802

0.803

0.612

0.826

 

Median

0.846

0.890

0.723

0.724

0.862

0.822

0.665

0.841

 

Q3

0.860

0.983

0.793

0.793

0.923

0.839

0.716

0.856

2

N

839

839

839

839

839

839

839

839

 

MIN

0.31

0.00

0.00

0.00

0.00

0.06

0.00

0.08

 

MEAN

0.65a

0.63c

0.50e

0.50e

0.64ab

0.58d

0.46f

0.64bc

 

MAX

0.85

1.00

1.00

1.00

1.00

0.79

1.00

0.82

 

STD

0.09

0.29

0.26

0.26

0.26

0.10

0.20

0.09

 

MSD

0.000

0.083

0.092

0.092

0.069

0.02

0.081

0.011

 

Q1

0.61

0.42

0.31

0.31

0.47

0.53

0.33

0.59

 

Median

0.66

0.64

0.48

0.48

0.68

0.59

0.47

0.65

 

Q3

0.71

0.91

0.67

0.67

0.85

0.64

0.59

0.70

3

N

1000

1000

1000

1000

1000

1000

1000

1000

 

MIN

0.81

0.60

0.61

0.61

0.69

0.80

0.54

0.78

 

MEAN

0.85a

0.72c

0.73c

0.73c

0.81b

0.85a

0.64d

0.81b

 

MAX

0.89

0.88

0.90

0.89

0.96

0.88

0.78

0.84

 

STD

0.01

0.04

0.04

0.04

0.04

0.01

0.04

0.01

 

MSD

0.0000

0.0193

0.0169

0.0176

0.0036

0.0002

0.0477

0.0017

 

Q1

0.85

0.69

0.70

0.70

0.79

0.84

0.61

0.81

 

Median

0.85

0.72

0.73

0.73

0.81

0.85

0.64

0.81

 

Q3

0.86

0.75

0.76

0.76

0.84

0.85

0.66

0.82

4

N

955

955

955

955

955

955

955

955

 

MIN

0.60

0.14

0.14

0.14

0.15

0.48

0.24

0.52

 

MEAN

0.72a

0.32e

0.33e

0.33e

0.36d

0.62c

0.63b

0.64b

 

MAX

0.79

0.52

0.53

0.52

0.56

0.72

0.93

0.72

 

STD

0.03

0.06

0.06

0.06

0.07

0.04

0.09

0.03

 

MSD

0.000

0.160

0.157

0.158

0.130

0.012

0.015

0.007

 

Q1

0.70

0.29

0.29

0.29

0.32

0.60

0.57

0.62

 

Median

0.72

0.32

0.33

0.33

0.37

0.62

0.64

0.64

 

Q3

0.74

0.36

0.37

0.37

0.41

0.65

0.70

0.66

  1. All the methods except 5 and 7 use cross-validation. M0 is the correlation between the predicted and the true simulated breeding values used as the benchmark for assessing the estimated predictive accuracy. N is the number of data sets out of a possible total of 1000 for which estimates were available for all the seven methods. Means for pairs of methods within each scenario with the same superscript letter are not significantly different at the 5% level of significance based on the t-test.
  2. MSD=Mean squared deviation, Q1 is the lower quartile and Q3 is the upper quartile. * The number of the equation used in the text is in parenthesis.