Skip to main content

Table 3 Results of the feature selection based on stepwise regression model regarding the grain yield as response variable for normal irrigation and drought stress conditions

From: Superoxide dismutase (SOD) as a selection criterion for triticale grain yield under drought stress: a comprehensive study on genomics and expression profiling, bioinformatics, heritability, and phenotypic variability

Normal irrigation

 Entered Variable

 

Removed variable

Partial R2

Model R2

F Value

p > F

 Protein

 

aNA

0.89

0.89

1395.98

0.017

Over all ANOVA of the final model of selected variables

 Source

 

DF

Sum of squares

Mean Square

F Value

p > F

 Model

 

1

881.368

881.36

52.67

<.0001

 Error

 

538

9002.63

16.73

NA

NA

Parameter estimation for the final model of selected variables

 Variable

Parameter

Standard parameter

Standard error

Sum of squares

F Value

p > F

 Intercept

1030.13

NA

4.12948

39,289

62,229.6

0.003

 Protein

−27.7947

−2.40241

0.74391

881.36

1395.98

0.017

Drought stress

 Entered Variable

 

Removed variable

Partial R2

Model R2

F Value

p > F

 SOD

 

NA

0.520

0.5195

73.02

<.0001

 Proline

 

NA

0.244

0.7635

28.23

<.0001

 GR

 

NA

0.042

0.8055

2.710

0. 003

Over all ANOVA of the final model of selected variables

 Source

 

DF

Sum of squares

Mean Square

F Value

p > F

 Model

 

3

2,380,092.0

793,364

35.99

<.0001

 Error

 

536

11,813,999

22,041.0

NA

NA

Parameter estimation for the final model of selected variables

 Variable

Parameter

Standard parameter

Standard error

Sum of squares

F Value

p > F

 Intercept

821.479

NA

40.6503

9,001,165

408.4

<.0001

 Proline

−4.14588

−0.173164

2.51802

59,751.00

2.710

0.030

 SOD

4.65863

0.347539

0.86511

639,160.0

29.00

<.0001

 GR

−8.43598

−0.102254

0.84753

2,183,721

99.08

<.0001

  1. a Not available: NA, glutathione reductase: GR, superoxide dismutase: SOD, probability level: p