Mondal et al., 2008. Livest. Res. Rural Dev., 20 (4): 63

Protein degradability by in sacco method and protein fractionation by laboratory analysis using CNCP (Cornell Net Carbohydrate and Protein) system was determined to develop a prediction equation for predicting degradability from different protein fractions.

Sixteen different types of feeds namely, maize grain (MZ), maize gluten meal (MGM), maize fibre (MF), maize oil cake (MOC), wheat (WT), wheat bran (WB), barley (BRL), soybean meal (SBM), sunflower meal (SFM), deoiled coconut cake (DCC), cottonseed cake (CSC), groundnut cake (GNC), mustard cake (MC), fish meal (FM), guar chuni (GC) and subabul leaves (LLM) were analysed for (A+B₁), B₂, B₃ and C. The above feeds were also incubated in the rumen of three fistulated, adult crossbred cattle to estimate the effective protein degradability (EPD) of individual feeds. A linear regression equation was drawn for EPD from different protein fractions.

Higher A+B₁ fraction was observed in MC, GNC and BRL, whereas lower values were noted in MGM, MF and MZ. B₂ fraction was highest in MF followed by MZ and LLM and lower values were found in GNC, MC and SBM. B₃ fraction was greater in MGM, FM and CSC, whereas the lower values were noted in MC, LLM and MF. Lignin bound nitrogen or C fraction was greater in MGM, GC and LLM and with lower values in BRL, WT, WB, MF and MZ. Higher EPD was found in MC, GNC followed by BRL, MOC and WB, whereas lower values were obtained in MGM, MF, LLM and CSC.

ECPD at different outflow rates were highly and positively correlated (r = 0.84 to 0.88) with (A+B₁) fraction, whereas with B₂, B₃ and C it was negatively associated (r = 0.55 to 0.62, 0.29 to 0.32 and 0.50 to 0.52 for B₂, B₃ and C, respectively). 

Significant predicted variables in stepwise regression included (A+B₁), (A+B₁)² and B₃² at fractional outflow rate of 0.02 h^-1 with R² value of 0.90. However, the predicted equations at fractional outflow rate of 0.04 and 0.06 h^-1 contained (A+B₁), (A+B₁)², B₃, B₃² and C² predictive variables with high degree of prediction (R² = 0.93). B₃ alone was weak predictor for UDP content of feeds.

Therefore, to avoid the maintenance of fistulated animals and to save time for in sacco experiment, laboratory analysis of protein fractions and fitting them in the regression equation may predict the EPD values effectively.