Establishment And Evaluation Of Effective Energy Prediction Model In Feeds Of Beef Cattle

Chemical analysis, digestion and metabolism trial, in vitro gas production experiments, in vitro enzymic experiments and in situ nylon bag experiments were conducted to evaluate the nutritional composition and effective energy values of 24 feedstuffs commonly used in ruminant feeding in China. Several prediction models were established through correlating calculation of 24 feedstuffs between their effective energy contents and the chemical components, in vitro gas production, in vitro enzymic degradation parameters and in situ degradation parameters.Expt.1 was conducted to measure the chemical compositions of 24 feedstuffs and compare the differences of chemical compositions among the feedstuffs. Feedstuffs in the experiment include corn stover, wheat stalk, rice straw, naked oats straw, soybean straw, oat hay, chinese wildrye grass, alfalfa, corn grain, barley grain, wheat grain, sorghum grain, soybean meal, cottonseed meal, peanut meal, flax meal, sunflower meal, distiller’s grains, beer brewers, pear pomace, apple pomace, soy sauce residue, distillers dried grains with soluble (DDGS), and corn bran. Dry matter (DM), organic matter (OM), ether extract (EE), ash, crude fiber (CF), neutral detergent fiber (NDF), acid detergent fiber (ADF), acid detergent lignin (ADL), neutral detergent fiber insoluble protein (NDIN), acid detergent insoluble nitrogen (ADIN), starch, calcium (Ca) and phosphorus (P) were measured. The results showed that nutrinent compositions differed significantly among different feedstuffs. Fiber content was higher for straws and hays, starch content higher for grains, CP content higher for oil seed meals, whilst chemical composition varied for other by-products. Similar results were observed in the present experiment with Feeding Standard of Beef Cattle of China (NY/T 815-2004) and Tables of Feed Composition and Nutritive Values in China (2004) for most feedstuffs.Expt.2. Nutrinent digestibilities and effective energy values of 24 common feeds were measured through in vivo experiments, and nutrinent digestibility and effective energy prediction models were established according to their chemical components. Twelve canuulated black Angus steers (initial body weight:317.3±42.8kg) were assigned into three groups based on their body weight. Each group consisted of 4 steers, and a 4×4 latin square experiment was conducted for each group. The experiment included two periods, each consisted of four phases, and 12 feedstuffs were measured at one period. The results showed that dry matter digestibility (DMD), organic matter digestibility (OMD), gross energy digestibility (GED), crude protein digestibility (CPD), digestible energy (DE) or metabolic energy (ME) had a positive relationship (P<0.01) with CP; whereas they were negatively related to crude fiber (CF), neutral detergent fiber (NDF), acid detergent fiber (ADF) and acid detergent lignin (ADL) (P<0.01). ADF was the most preferable predictor for single factor prediction equations, whilst multi-factor models based on the composition of the fiber and nonfiber fractions were more accurate. Precision and accuracy improved while the number of predictors increased. DE and digestible organic matter (DOM) had a positive relationship (P<0.01) with ME, and the correlation equations were ME (MJ/kg DM)=0.885DE (MJ/kg DM) (R2=0.950, n=96, P<0.01) and ME (MJ/kg DM)=0.224DOM (%DM)-1.586 (R2=0.881, n=96, P<0.01). In summary:prediction equations for nutrinent digestibilities and effective energy values can be accurately established according to chemical compositions of the feedstuffs.Expt.3. A total of 24 feedstuffs were analyzed for in vitro gas production, gas dynamic and fermentation characteristics using in vitro gas production method. Nutrinent digestibility and effective energy prediction models were established according to the data obtained from the present experiment combined with in situ experiments and chemical compositions of the feeds. Three canuulated black Angus steers (bodyweight:380±15kg) were used. The results showed that in vitro gas production, gas dynamic and fermentation characteristics differed significantly among different feedstuffs. Nutrinent digestibilities or effective energy values had a positive relationship (P<0.01) with gas production (GP), calculated maximal GP (B), fractional rate of GP (c) or total valiate fatty acid concentration (TVFA) of 24h fermentation fulid; whereas they were negatively related (P<0.01) to the acetic propotion. Correlation coefficient (r) between nutrinent digestibility and GP decreased as the incubation time became longer, and effective energy and GP as well. GP12h, or TVFA was the most preferable predictor for single factor prediction equations, whilst higher determination coefficient (R2) values for multi-factor models combined with acetic propotion value and chemical composition were observed. In addition, precision and accuracy improved while the number of predictors increased. In summary: prediction equations for nutrinent digestibilities and effective energy values can be accurately established according to the results of in vitro gas production experiment, and accuracy of the equations were further improved combined with the chemical composition results of the feedstuffs.Expt.4. A total of 24 feedstuffs were analyzed for dry matter solubility (DMS) and organic matter solubility (OMS) by using in vitro pepsin-cellulose (PC) enzymatic method and neutral detergent-cellulose (NC) enzymatic method. Nutrinent digestibility and effective energy prediction models were established according to the data in the present experiment combined with in situ experiments of the feeds. The results showed that DMS vaules of the two enzymatic methods were significantly correlated (P<0.01, r=0.963), as well as their OMS values (P<0.01, r=0.957). However, DMSNC and OMSNC were significantly higher (P<0.01) than DMSPC and OMSPC. Moreover, precision and accuracy of prediction equations were higher using DMSNC or OMSNC as predictors than DMSPC or OMSPC. Correlation coefficient (r) of DMSNc and OMSNC was 0.998 (P<0.01), whilst r=0.999 for DMSPC and OMSPC (P<0.01). In addition, similar results were observed using DMS or OMS, which were measured by the same enzymatic method. In summary:prediction effects were different using different enzymatic methods to establish nutrinent digestibility and effective energy prediction models; compared with pepsin-cellulose enzymatic method, the prediction effect was better using neutral detergent-cellulose enzymatic method.Expt.5. Twelve feedstuffs were measured for ruminal disapperance rate and degradation parameters by using the nylon bag method. Nutrinent digestibility and effective energy prediction models were established according to the data in the present experiment. Three canuulated black Angus steers (bodyweight:380±15kg) were used. Feedstuffs include corn stover, rice straw, soybean straw, chinese wildrye grass, beer brewers, sunflower meal, distillers dried grains with soluble (DDGS), soybean meal, corn grain, barley grain, wheat grain and sorghum grain. The results showed that in situ ruminal dry matter disappearance rate (ISDD) and neutral detergent fiber disappearance rate (ISND) and degradation parameters differed significantly among different feedstuffs (P<0.01). Dry matter digestibility (DMD), organic matter digestibility (OMD), digestible energy (DE) or metabolic energy (ME) had a positive relationship (P<0.01) with ISDD of different incubation times, and correlation coefficient (r) increased as the incubation time became longer. DMD, OMD, DE or ME also had a positive relationship (P<0.01) with rapid degraded fraction (a), rapid degraded fraction+slowly degraded fraction (a+b) or effective degradability of dry matter (DM), however, their relationship with degradation rate (c) of DM or ISND was not significant (P>0.05). ISDD72h was the most preferable predictor for single factor prediction equations (R2>0.8), whilst higher determination coefficient (R2>0.85) values for multi-factor models combined (a+b) with c as predictors were observed. In summary:prediction equations for nutrinent digestibilities and effective energy values can be accurately established according to the results of in situ nylon bag experiment, and the most preferable predictors are (a+b) and c.