Relationship between amino acid scores and protein quality indices based on rat growth.

Protein efficiency ratio (PER), relative PER (RPER), net protein ratio (NPR) and relative NPR (RNPR) values, and amino acid scores were calculated for 20 food products (casein, casein + Met, beef salami, skim milk, tuna, chicken frankfuters, sausage, heated skim milk, peanut butter, rolled oats, soy isolate, chick peas, pea concentrate, kidney beans, wheat cereal, pinto bean, lentils, rice-wheat gluten cereal, macaroni-cheese, and beef stew). In most cases, PER, RPER, NPR or RNPR ranked the products in the same order and positive correlations among the protein quality methods were highly significant (r = 0.98-0.99). Amino acid scores (based on the first limiting amino acid, Lys-Met-Cys, Lys-Met-Cys-Trp or lys-Met-Cys-Trp-Thr) were positively correlated to the PER, RPER, NPR or RNPR data (r = 0.61-0.75). Inclusion of the correction for true digestibility of protein improved the correlations between amino acid scores and the indices based on rat growth. The correlations were especially high between Lys-Met-Cys scores (corrected for true digestibility of protein) and PER, RPER, NPR or RNPR (r = 0.86-0.91). Inclusion of the correction for true digestibility of individual amino acids did not result in further improvements of the correlations in most cases. It is concluded that adjusting amino acid scores for true digestibility of protein would be sufficient and further correction for digestibility of amino acids would be unnecessary in mixed diets.