Skip to main content

Table 4 The response surface methodology (RSM) yield-vigor model, the artificial neural network(ANN) yield model, the vigor model, and the viable cell number model predictions were validated, and the results were considered credible with a relative error of less than 5%

From: Optimization of preparation conditions for Salsola laricifolia protoplasts using response surface methodology and artificial neural network modeling

Model

Cellulase R-10 content (%)

Macerozyme R-10 content (%)

Mannitol concentration(mol L−1)

Protoplast yield(× 106/100 mg)

Protoplast viability (%)

Number of viable cells(× 106/100 mg)

Relative error (%)

Predicted value

Experiment value

Predicted value

Experiment value

Predicted value

Experiment value

Protoplast yield

Protoplast viability

Number of viable cells

RSM yield-viability

1.86

1.00

0.50

1.464

1.470

90.81

89.73

1.329

1.319

0.41

1.20

0.76

ANN-yield

1.98

1.00

0.50

1.625

1.550

90.16

90.65

1.337

1.405

4.84

0.54

4.83

ANN-viability

1.00

0.75

0.50

0.565

0.563

90.18

90.81

0.510

0.508

0.36

0.69

0.39

ANN-number of viable cells

2.00

1.00

0.70

1.574

1.566

87.12

87.27

1.342

1.367

0.51

0.17

1.83