Table 12.
Dose response regressions for Chinese Yellow broiler breeder hens fed diets with different copper content.
| Variable | Model | Regression equation1 | Optimal dietary Cu level, mg/kg | Optimal daily Cu fed allowance, mg | P-value | R2 |
|---|---|---|---|---|---|---|
| MDA2 in plasma, nmol/mL | QP2 | Y = 5.65 − 0.128 × X − 1.77 × 10−3 × X2 | 36.2 | 4.16 | 0.002 | 0.438 |
| Two-slope BL3 | Y = 6.77 − 0.261 × X (X ≤ 17.2) Y = 1.06 + 0.071 × X (X > 17.2) |
17.2 | 1.98 | <0.001 | 0.561 | |
| Copper in liver, mg/kg DM | BL with plateau4 | Y = 15.0 (X ≤ 21.2) Y = 11.2 + 0.180 × X (X > 21.2) |
21.2 | 2.44 | <0.001 | 0.636 |
| Cu-ATPase in liver, μmolPi/mg protein/hour | BL with plateau4 | Y = 5.66 (X ≤ 15.7) Y = 5.16 + 0.032 × X (X > 15.7) |
15.7 | 1.81 | <0.001 | 0.706 |
| Shell thickness, mm | QP2 | Y = 0.344 + 9.04 × 10−4 × X − 1.17 × 10−5 × X2 | 38.6 | 4.44 | 0.001 | 0.458 |
| Two-slope BL3 | Y = 0.348 + 4.20 × 10−4 × X (X ≤ 41.1) Y = 0.390 – 6.20 × 10−4 × X (X > 41.1) |
41.1 | 4.73 | <0.001 | 0.453 |
Abbreviation: MDA, malondialdehyde.
Regression equations obtained using the analyzed Cu in the trial diets (3.5, 8.5, 13.5, 23.5, 43.5 and 83.5 mg/kg).
QP: Quadratic polynomial; QP model: Y = α + β × X + γ × X2, where Y is the response variable, X is the dietary Cu, α is the intercept; β and γ are the linear and quadratic coefficients respectively. The optimal response was obtained by–β/(2 × γ).
BL: Broken line; 2-slope BL model: Y = α + β × Cu, Cu ≤ γ; Y = δ + ε × Cu, Cu > γ, where Y is the response variable, X is the dietary Cu, both α and δ are intercepts, both β and ε are slopes of lines. The Cu level at the break point (γ) was considered as the one providing the optimal response.
BL with plateau model: Y = α + β × γ, Cu ≤ γ; Y = α + β × Cu, Cu > γ, where Y is the response variable, X is the dietary Cu, α is the intercept, β is the slope of line, the value (α + β × γ) is the plateau. The Cu level at the break point (γ) was considered to be that providing the optimal response.