Supplemental material for "Evaluation of ruminal outflow of protein and nitrogen fractions, and total and individual essential amino acids predictions by nutritional models in dairy cattle"

The objective was to evaluate the fit statistics for the predictions of ruminal individual EAA outflows by NRC (2001), NASEM (2021), CNCPS (v.6.5.5) and NittanyCow (NC) models in lactating dairy cows. Sixty-seven studies published in English between 1984 and 2020 with 252 treatment means were considered for analysis. Lactational performance (DMI, DIM, milk yield, milk fat and true protein, and BW data), dietary nutrient composition (CP, NDF, ether extract, and ash), and ruminal outflow of total AA, microbial N, nonammonia nonmicrobial N, and individual EAA determined by omasal or duodenal digesta sampling techniques were extracted from publications and used in the analysis when available. Production and dietary nutrient composition data reported in the studies were used as inputs to predict ruminal outflows of total AA, NAN, microbial protein and N, nonammonia nonmicrobial CP and N, and individual EAA using NRC, NASEM, CNCPS and NC models. For all nutritional models, dietary concentration (% DM) of CP was identical or allowed to vary by 0.5%-units maximum to that reported in the studies (e.g., 16.0 versus 16.5% CP on a DM basis for observed and predicted values). Fit statistics and performances of nutritional models were assessed using Lin’s concordance correlation coefficient (CCC), root mean squared error (RMSE, g/d and % of observed), as well as mean and linear biases. Models performed similarly when predicting ruminal total AA outflow with CCC ranging from 49 to 57% and RMSE ranging from 19 to 30% of observed. Predictions of ruminal MicP and NANMCP outflows had CCC ranging from 38 to 60% and RMSE ranging from 24 to 40% of observed. Predictions of Lys, Met, and His had CCC ranging from 38 to 62% and RMSE ranging from 23 to 40% observed. Overall, model fit statistics indicated superior performance of both NRC, NASEM, and NC in predicting all individual EAA outflows compared with CNCPS. NittanyCow and NASEM, respectively, demonstrated superior performance in predicting total AA while CNCPS was the most accurate in predicting NAN and NANMCP (i.e., NANMN × 6.25) outflows. Mean and observed biases of concern were identified across all models (mean or linear bias > 5.0% of observed), and these biases should be considered by nutritionists when balancing and evaluating rations targeting individual EAAs in high-producing dairy cows.