Table 2 Prediction statistics of fatty acid profile of intact fat.

From: Prediction of fatty acid composition in intact and minced fat of European autochthonous pigs breeds by near infrared spectroscopy

Parameter

Spectrum range (cm−1)

Math. treat

nPLS

Calibration

Cross validation

External validation

R2

RMSE

R2cv

RMSEcv

RPDcv

R2v

RMSEv

RPDv

RER

C12:0

7500–5400

SNV; DT; SG

14

0.3262

0.0121

0.2402

0.0129

1.16

0.2150

0.0133

1.18

6.97

C13:0

7500–5400

SNV; DT

13

0.1505

0.0010

0.1042

0.0010

1.06

0.1040

0.0045

1.02

9.76

C14:0

7500–5400

SNV; DT; SG

13

0.3573

0.1683

0.2648

0.1806

1.18

0.1704

0.2059

1.11

6.53

C15:0

7500–5400

SNV; DT; SG

15

0.1501

0.0136

0.1170

0.0143

1.02

0.2167

0.0107

1.10

7.18

C16:0

7500–5400

SNV; DT

15

0.7204

1.1640

0.6617

1.2840

1.72

0.6820

1.2327

1.78

9.05

C17:0

7500–5400

SNV; DT

15

0.1970

0.0816

0.1080

0.0860

1.06

0.1014

0.0827

1.03

5.81

C18:0

7500–5400

SNV; DT

14

0.7351

0.9458

0.6927

1.0210

1.80

0.6970

1.0080

2.10

14.25

C20:0

7500–5400

SNV; DT

13

0.3013

0.0359

0.2332

0.0380

1.13

0.2622

0.0369

1.19

7.24

C22:0

7500–5400

SNV; DT

14

0.1590

0.0114

0.1500

0.0121

1.03

0.1490

0.0109

0.92

4.56

Total SFA

7500–5400

SNV; DT

15

0.8586

1.3512

0.8356

1.4610

2.46

0.8081

1.6320

2.29

14.24

C14:1

4500–9500

SNV; DT

13

0.3963

0.0045

0.3208

0.0047

1.24

0.3142

0.0040

1.46

6.79

C16:1

4500–9500

SNV; DT; SG

14

0.4649

0.4306

0.4018

0.4567

1.29

0.5258

0.4177

1.46

5.93

C17:1

4500–9500

SNV; DT; SG

15

0.3070

0.0881

0.2382

0.0927

1.14

0.3327

0.0825

1.23

7.38

C18:1

4500–9500

SNV; DT; SG

14

0.8744

1.3190

0.8517

1.4374

2.59

0.7714

1.6598

2.10

11.95

C20:1

4500–9500

SNV; DT; SG

14

0.3171

0.1710

0.2452

0.1807

1.15

0.2657

0.1747

1.17

6.71

C22:1 n9

4500–9500

SNV; DT; SG

14

0.3405

0.0153

0.2821

0.0160

1.18

0.3241

0.0181

1.20

4.34

Total MUFA

4500–9500

SNV; DT; SG

14

0.8853

1.3406

0.8660

1.4535

2.73

0.7812

1.7164

2.15

12.24

C18:2 n6

5400–6100/7400–8400

SNV; DT

13

0.9581

0.5963

0.9444

0.6892

4.23

0.9054

0.8876

3.26

14.63

C18:3 n3

5400–6100

SNV; DT

13

0.7988

0.1169

0.7411

0.1324

1.97

0.7988

0.1224

1.92

9.22

C20:2 n6

5400–6100/7400–8400

SNV; DT

14

0.7905

0.0740

0.7671

0.0790

2.07

0.7773

0.0808

2.13

8.74

C20:3 n3

5400–6100

SNV; DT

14

0.6257

0.3390

0.5553

0.0370

1.50

0.5441

0.0368

1.49

9.70

C20:3 n6

5400–6100/7400–8400

SNV; DT

14

0.5163

0.0104

0.4702

0.0109

1.38

0.5505

0.0101

1.51

8.37

C20:4 n6

5400–6100/7400–8400

SNV; DT

14

0.5752

0.0233

0.5364

0.0244

1.47

0.5817

0.0233

1.56

7.56

C20:5 n3

5400–6100

SNV; DT

14

0.4999

0.0017

0.4029

0.0018

1.35

0.3666

0.0019

1.26

7.46

C22:4 n6

5400–6100/7400–8400

SNV; DT

14

0.4262

0.0130

0.3782

0.0136

1.27

0.4358

0.0137

1.35

7.56

C22:5 n3

5400–6100

SNV; DT

14

0.5113

0.0100

0.4072

0.0118

1.30

0.5143

0.0107

1.45

7.70

Total PUFA

5400–6100

SNV; DT

14

0.9512

0.7450

0.9453

0.7913

4.27

0.9265

0.9071

3.70

16.36

n3 PUFA

5400–6100

SNV; DT

14

0.7967

0.1445

0.7312

0.1666

1.93

0.7256

0.1537

1.91

9.73

n6 PUFA

5400–6100/7400–8400

SNV; DT

14

0.9429

0.7424

0.9344

0.7983

3.90

0.9054

0.9522

3.26

14.41

  1. SNV standard normal variate, DT de-trending, SG Savitzky–Golay filter, nPLS number of partial least square terms, R2c coefficient of determination in calibration, RMSEC root mean square error of calibration, R2cv coefficient of determination in cross-validation, RMSEcv root mean square error of cross validation, RPDcv residual prediction deviation in cross validation, R2v coefficient of determination of external validation, RMSEv root mean square error of external validation, RPDv residual prediction deviation in external validation, RER range error ratio in external validation, SFA saturated fatty acids, MUFA monounsaturated fatty acids, PUFA polyunsaturated fatty acids.
Back to article page