FOOD AND AGRICULTURE ORGANIZATION OF THE UNITED NATIONS	ESN: FAO/WHO/UNU EPR/81/31 August 1981
	WORLD HEALTH ORGANIZATION
	THE UNITED NATIONS UNIVERSITY

Provisional Agenda Item 3.2.3

Joint FAO/WHO/UNU Expert Consultation on
Energy and Protein Requirements

Rome, 5 to 17 October 1981

AMINO ACID SCORING PATTERNS

by

A. Harper
University of Wisconsin
USA

This paper represents an effort to explain the differences between the amino acid scoring patterns for evaluating proteins proposed by the FAO/WHO Expert Committee that met in 1971 (1) and the Food and Nutrition Board Committee on Amino Acids (USA) which considered this subject as part of its project on assessment of the merit of amino acid fortification during the period between 1966 and 1971 (2). Both Committees used the same major sources of information about amino acid requirements of infants, adolescents and adults but there were a few differences in the way these were used and in the way they were supplemented with information about protein consumption and the amino acid composition of milk and other foods. As is indicated below, it was mainly the differences in the judgments of the Committees and in the sources of supplemental information that led to the somewhat different scoring patterns proposed by the two Committees.

The approach taken in dealing with this subject is to compare the estimates of amino acid requirements of adults, adolescents and infants as reported by the two Committees, pointing out differences in the values and, wherever it is possible, explanations for these. The provisional scoring patterns will then be compared and an effort will be made to identify reasons for differences between them and to assess the significance of these differences.

Amino Acid Requirements of Adults

Both Committees compiled values for amino acid requirements of adults from the studies of men by Rose and associates (3); of women by Leverton and associates (4); and from a few other studies that supplemented these (5–8), together with an analysis of the available information by Hegsted (9).

The values compiled by the two Committees are presented in Tables 1 and 2. Table 1 is from the FAO/WHO report (Table 17, p. 55). Table 2 is from the NRC report (Table 6, p. 40). The values selected for men by the two Committees (column 1 of Table 1 and column 5 of Table 2) are identical except for the requirement for total sulfur-containing amino acids for which the NRC Committee selected the lower value of 1010 mg/day obtained by Rose and Wixom (6) when they included both methionine and cystine, rather than methionine alone, in the experimental diets of their subjects. The values for women (column 2 of Table 1 and column 1 of Table 2) selected by the NRC Committee were the highest values for any individual, to conform with the procedure used by Rose. Thus the NRC value of 700 mg/day for total sulfur-containing amino acids exceeds that of 550 mg/day selected by the FAO/WHO Committee. The higher values selected by the FAO/WHO Committee for leucine and lysine requirements of females represent differences in the judgments of the two Committees as to the highest individual value for achievement of nitrogen equilibrium. Values for the amino acid requirements of women from the re-evaluation published by Hegsted (9) were included in both reports and (column 3 of both Tables 1 and 2) are identical.

^a Taken from Rose.¹²⁰ The values represent the highest estimate of individual requirement to achieve positive nitrogen balance.
^b Taken from Leverton, Swendseid, Jones, Reynolds, Clark, Linksweiler, Burrill and their coworkers (the references have been summarized by Irwin & Hegsted ¹²¹). The values represent the highest estimate of individual requirement to achieve the zone of nitrogen equilibrium (balance of0 = 5% of intake).
^c Data of some of the above authors recalculated by Hegsted ¹²² using regression analysis to estimate the average requirement to achieve nitrogen equilibrium.
^d Derived estimate emphasizing the upper range of individual requirements.
^e Assuming a safe level of protein intake of 0.55 g per kg per day (averaged value for men and women).
*Table 17, FAO/WHO (1), p. 55.

^cLeverton, 1959.
^dHegsted, 1963.
^eRose, 1957.
^fReynolds et al. 1958.
^gRose and Wixom, 1955c. Cys will spare 80–89 percent of the Met requirement.
^hBurrill and Schuck, 1964.
ⁱRose and Wixom, 1955c. Tyr will spare 70–75 percent of the Phe requirement.
^jTSAA is total sulfur-containing amino acids; TAAA is total aromatic aminoacids.
* From Table 6, NRC (2), p. 40.

The FAO/WHO Committee averaged the adult values expressed as mg/kg of body wt/day and derived an amino acid scoring pattern for adults (column 5 of Table 1). This was based on the assumption that the amounts of individual amino acids required (column 4 of Table 1) should be provided by 0.55 g of protein, the estimated safe level of intake of high quality protein. The NRC Committee did not propose an amino acid scoring pattern for adults. It did, however, select the set of requirements for adults estimated from the re-evaluation by Hegsted, as probably the most reliable set of values after reviewing the entire published data. Both Committees concluded that the apparent sex difference between requirements of males and females was not significant.

Differences between the two sets of requirements (columns 4 of Tables 1 and 2) are not great. The NRC estimate for the lysine requirement is low; the higher FAO estimate is largely attributable to the particularly high value for males from the experiments of Rose in which the highest value for a single subject was accepted as the requirement. The somewhat higher estimates for leucine and total aromatic amino acid requirements in the FAO/WHO report are similarly explained.

Both Committees pointed out that the estimates of amino acid requirements are from nitrogen balance studies and did not take into account integumental and other minor nitrogen losses or the additive errors of the balance technique, so probably underestimate somewhat the actual requirements. Values calculated by Hegsted to be required for positive nitrogen retention of 0.5 gm/day were much higher than the values determined from regression analysis of requirements for nitrogen equilibrium. They were considered, by the NRC Committee, to be unrealistic as there were far too few values at the upper end of the requirement range to permit identification of an inflection point in the regression line on which the analysis was based.

Amino Acid Requirements of Children

Both Committees compiled values for amino acid requirements of children, aged 10–12 years, from the publications of Nakagawa et al. (10–15). The FAO/WHO Committee accepted values from the publications of Nakagawa et al. for the upper range of individual requirements for achievement of positive nitrogen balance (Table 3). The NRC Committee recognized that Nakagawa et al. had used rather large increments of amino acids in estimating requirements, that some of the values for requirements were not proportional to either the adult or infant requirements and that Snyderman (16) had observed a decline in phenylalanine requirement by 18 months to less than one-third of the requirement at birth (Figure 1). The NRC Committee, therefore, estimated requirements for children from the information about individual subjects given in the published reports (Table 4). These values are, without exception, lower than those accepted by the FAO/WHO Committee. The validity of this approach may be open to question; nevertheless, the values, with the exceptions of the requirements for phenylalanine and tryptophan which seem inordinately low, are more in line with those for the infant if it is assumed that other requirements fall at about the same rate as that for phenylalanine (16).

^a Based on Nakagawa et al.^125–126 The values represent estimates of the upper range of individual requirements for the achievement of positive nitrogen balance in boys.
^b Based on a safe level of protein intake of 0.8 g per kg per day, the average of safe levels of protein for boys and girls in that age group.
* Table 18, FAO/WHO (1), p. 56.

^b Nakagawa et al., 1960, 1961a,b, 1962, 1963, 1964.
* From Table 6, NRC (2), p. 40.

FIGURE 1. The phenylalanine requirement of the phenylketonuric infant related to age; average of 50 infants and one standard deviation above and below the average (Snyderman, 1974).

The FAO/WHO Committee calculated an amino acid scoring pattern for children based on the assumption that the safe level of intake of high quality protein for this age group is 0.8 gm/kg of body wt/day. It is noteworthy that the values per gm of protein for lysine and total sulfur-containing amino acids in this pattern are higher than those for the infant.

Amino Acid Requirements of Infants

Both Committees tabulated the amino acid requirements of infants, as estimated by Holt and Snyderman (17). The values in column one of Tables 5 and 6 are in agreement with the exception of that for leucine. The lower value reported by the NRC Committee for the leucine requirement was provided by Dr. L. E. Holt, Jr. from a study done after the earlier report used by the FAO/WHO Committee.

Both Committees also estimated amino acid requirements of infants from the amounts of protein consumed by infants fed on formulas and who were growing well in studies done in the Pediatrics Department at the University of Iowa (18, 19). The values reported by the two Committees are quite different (column 2 in Table 5 and column 4 in Table 6). The FAO/WHO Committee used values estimated by Foman and Filer (19). The NRC Committee calculated intakes of amino acids of infants fed three different formulas from the published nitrogen intakes (18) using values for amino acid composition of foods compiled by Orr and Watt (20). The lowest value from among the three estimates was accepted as the requirement. With the exception of the requirement for tryptophan, the values estimated from the amounts of protein consumed in formulas were lower than those estimated by Holt and Snyderman. It should be recognized that infants in the studies by Holt and Snyderman were as young as two months of age whereas those studied by Foman and associates were 4.5 to 6 months of age.

^a Requirements estimated when amino acids were fed or incorporated in basal formulas. The values represent estimates of maximal individual requirements to achieve normal growth. ^123
b Calculated intakes of amino acids when formulas were fed in amounts sufficient to maintain good growth in all the infants studied; the amino acids were not varied independently. ^124
c Based on a safe level of intake of 2 g protein per kg per day, the average of suggested levels for the period 0–6 months.
^d The values for cystine and tyrosine were estimated on the basis of the methionine: cystine and phenylalanine: tyrosine ratios in human milk (see Table 20).
* Table 18, FAO/WHO (1), p. 56.

^aHolt and Snyderman, 1965.
* From Table 6, NRC(2), P. 40.

*From Table 8, NRC (2), p. 46.

With respect to the differences in the estimated requirements of infants consuming formulas reported by the two Committees, it should be noted that a later set of requirement values reported by Fomon (21) are much more in line with those of the NRC Committee. A comparison of the earlier values and the later ones is given in Table 8.

It should also be noted that Snyderman and associates (22) were able to maintain satisfactory rates of growth in a small group of infants who consumed a cow's milk formula that was diluted and to which was added glycine and urea. These infants were consuming amounts of amino acids slightly below those estimated to be adequate from the studies of Foman and associates (column 4 of Table 7).

Amino Acid Scoring Patterns

The amino acid scoring patterns for different age groups suggested by the FAO/WHO Committee are presented in Table 9 together with patterns proposed by earlier FAO/WHO Expert Committees (23, 24). Similar information from the NRC Committee Report is presented in Table 10.

In assessing the earlier amino acid scoring patterns, it should be recalled that information about amino acid requirements of infants had not been published in 1957 when the first FAO scoring pattern was proposed. This pattern was based mainly on the amino acid requirements of adults. The values are essentially double those of the pattern proposed for adults by the FAO/WHO Committee in 1973. It became evident subsequently, particularly from observations that infants consuming cow's milk formula grew well, that the value for sulfur-containing amino acids in the 1957 pattern was high.

The 1965 FAO/WHO Expert Committee did not propose an amino acid scoring pattern based on studies of amino acid requirements but, instead, proposed that the amino acid patterns of human milk, cow's milk or whole egg could serve this purpose. The agreement among the values for the amino acid compositions of these proteins reported in the 1973 FAO/WHO and 1974 NRC reports is striking, considering that the values were from different sources. The low value for the total aromatic amino acid content of human milk in the FAO/WHO report (column 6 in Table 9) is undoubtedly a transcription error and should be 92 rather than 72. The value for the isoleucine content of cow's milk (column 7 of Table 9) appears to be low. The one disturbing feature is the range of values reported by the FAO/WHO Committee for the amino acid composition of human milk. This would suggest that human milk proteins can range between 40 and 60% in their content of essential amino acids. The lower end of this range would be close to that expected in plant proteins; the upper end would exceed that of animal proteins generally.

^a See preceding tables.There is no evidence for or against a histidine requirement for young children.
^b Compositions reported by FAO,¹⁴ Lindner et al.,³³ and Soupart et al.^129
c Composition reported by FAO;¹⁴ value for tryptophan by microbiological assay.
^d Composition reported by Lunven et al.¹³⁶
*Table 20, FAO/WHO, p. 58.

^aOrr and Watt, 1968.
* Table 11, NRC, p. 56.

The proposal that the amino acid compositions of these proteins could serve as guides for amino acid scoring of proteins posed new problems. The value for total sulfur-containing amino acids in cow's milk is well below that of the 1957 scoring pattern but the value for egg proteins greatly exceeded that of the 1957 pattern. Also, the values for lysine for all three proteins were well above that of the 1957 pattern, particularly those of cow's milk and egg.

The efforts of the most recent FAO/WHO Expert Committee and the NRC Committee to remedy these problems culminated in proposals for new amino acid scoring patterns. That proposed by the NRC Committee is presented in the last column of Table 10 and that proposed by the FAO/WHO Committee is presented in Table 11. Both Committees concluded that the scoring patterns should be designed for evaluating the adequacy of proteins for the young child and that they should be based mainly on the estimated amino acid requirements of infants. The proposed patterns are expressed as mg of amino acid per gram of protein. The amino acid values were selected so that a protein having an amino acid composition matching or exceeding the scoring pattern should meet the requirements for all of the essential amino acids when it is fed in an amount that will meet the total requirement for high quality protein.

*Table 21, FAO/WHO, p. 63

The approach used by the NRC Committee was to select for the scoring pattern values per gram of protein that were about half those of the infant amino acid requirements per kg of body weight. Two grams per kg of such a protein, the estimated requirement of the infant for high quality protein, would thus provide enough of each of the essential amino acids to meet the estimated amino acid requirements of the infant. Then, as amino acid requirements fall more rapidly with age than the protein requirement, such proteins should be more than adequate in essential amino acid content for other age groups.

The FAO/WHO Committee used essentially the same procedure but took into account the estimated amino acid requirements for both infants and young children. It also considered the amounts of amino acids consumed from foods by young children who were growing normally. It also rounded off all of the values, except that for tryptophan, to the nearest 5 mg.

As the result of these somewhat different procedures, there are some differences in the proposed patterns. The values for isoleucine, leucine, tryptophan and valine are practically the same in both patterns. The value for lysine is about 8% higher in the FAO/WHO pattern; this is the result of using the slightly higher estimate of the requirement for lysine from Holt and Snyderman and rounding off the value to the next highest 5 mg. The value for total aromatic amino acids in the FAO/WHO scoring pattern is 18% below that of the NRC pattern. This is attributable to the lower value for the requirement for total aromatic amino acid requirements of infants selected by the FAO/WHO Committee from the review article by Foman and Filer whereas the NRC Committee selected the higher values calculated from intakes by infants fed three different formulas (compare column 2 in Table 5 with column 4 in Table 7). The value for threonine in the FAO/WHO pattern is 14% higher than that in the NRC pattern. This is attributable to use of the higher value for the threonine requirement of infants reported by Holt and Snyderman by the FAO/WHO Committee as compared to the lower value from estimates of threonine intakes by infants fed various formulas and growing well selected by the NRC Committee (compare column 1 in Table 5 with column 4 in Table 7).

The value for total sulfur-containing amino acids shows the greatest discrepancy. The FAO/WHO value is 34% higher than that selected by the NRC Committee. The FAO/WHO Committee used the estimated requirement for young children as the basis for this value rather than the infant requirement which was lower per gram of protein (compare the last columns of Tables 3 and 5). The NRC Committee concluded that this value was high on the basis of a re-evaluation of the information in the original paper.

The major difference between the two amino acid scoring patterns that is of some concern is that between the values for methionine plus cystine. This subject has been reviewed (25, 26) since the two committee reports were published and some additional information (27–29) on amino acid requirements of young children that bears both specifically on the requirement for methionine plus cystine and also on amino acid requirements more generally, has since become available.

It was pointed out above that, in establishing the pattern of amino acid requirements of infants, the FAO/WHO Expert Committee used values for amino acid intakes of infants published in a review article in 1967 by Fomon and Filer (19) and that these did not agree well with values published subsequently in 1974 by Fomon (21). Harper (26), therefore, recalculated the values for amino acid intakes of infants in the studies by Fomon and associates, using the FAO/WHO values for protein intakes by infants fed cow's milk formulas (Table 15 in the 1973 FAO/WHO report) and the FAO/WHO values for amino acid composition of milk as given in Table 9 (Table 20 of the FAO/WHO Report). The results of these calculations are shown in Table 12. If the recalculated values based on protein intakes (given in parentheses in column 3 of Table 12) are substituted for the respective values in the FAO/WHO estimates of requirements (Table 12, column 3), and also in the FAO/WHO proportionality pattern (Table 12, column 5), differences between the NRC and FAO/WHO tabulations would be small. It should be noted that the proportionality patterns in Table 12 are not the final amino acid scoring patterns selected by either the NRC Committee or the FAO/WHO Committee.

Fomon et al. (27) have published information on methionine intakes of infants who grew well while they were being fed a soybean protein formula with or without a supplement of methionine. These infants were observed over the period of growth from 8 to 111 days of age. They were thus younger than most of the infants in other studies of amino acid requirements. The intakes of total sulfur-containing amino acids for the period between 42 and 111 days of age were, for the supplemented group 57.6 mg/100 kcal and for the unsupplemented group 50.7 mg/100 kcal. The two groups consumed 102 and 106 kcal/kg/day, respectively, during this period. Nitrogen retentions of the two groups were similar. The methionine-supplemented group had slightly higher serum albumin and slightly lower serum urea concentrations than the unsupplemented group during the early part of the experiment. What is noteworthy is that these young infants who were growing well had methionine plus cystine intakes slightly below the FAO/WHO, and slightly above the NRC, estimated requirements for infants 4.5 to 6 months of age. The requirements for the older infants should be lower, as requirements fall with increasing age. This comparison would seem to indicate, however, that the requirement for the sulfur-containing amino acids of infants falls less rapidly than that for the aromatic amino acids with increasing age. The same would be true for older age groups (2) if the estimates of requirements for sulfur-containing amino acids of adolescents (Tables 3 and 4) and adults (Tables 1 and 2) are accepted (see also column 5 of Table 13, below).

Quite recently information about amino acid requirements of 2-year old children has been reported from INCAP (28, 29). The children studied were fed a diet containing a low level of protein with additional nitrogen from an amino acid mixture so the dietary level of each amino acid in turn could be altered readily while total “protein” intake was kept constant. Growth, nitrogen retention and several biochemical indicators were monitored. From measurements of amino acid intakes to maintain satisfactory performance, estimates of amino acid requirements were made. These (28, 29) are given in column 2 of Table 13 together with the NRC estimates of requirements per kg of body weight per day for other age groups (Tables 2,4 and 6).

¹ Assuming protein requirement is 2 gm/kg body wt/day.
² FAO/WHO report (2).
³ Figures in parentheses calculated from average values for milk composition in Table 20 of FAO/WHO report (2).
^* Table 2 from reference 26.

From these estimates, the amino acid requirements, with the exceptions of those for lysine and tryptophan, would appear to decline by about 50% between 6 months and 2 years of age. This is about the same rate of decline (Figure 1) as was reported by Snyderman (16) for the phenylalanine requirement and is somewhat greater than the estimated decline in the protein requirement (1, 2).

The question that still remains to be settled is: “What is the appropriate amino acid scoring pattern for evaluating dietary proteins?” The proposed scoring patterns are listed in Table 14. The INCAP values have been calculated per gram of protein from the information on amino acid requirements presented by Pineda et al. (28).

The values derived from the INCAP studies for valine and isoleucine are considerably lower than those proposed by either Committee; the values for lysine and tryptophan do not deviate greatly from those of the provisional patterns. As the INCAP values are based on studies of two-year olds who should have lower amino acid requirements than younger children per unit of body weight, although not necessarily per gram of protein, it would seem unwise to lower the values for valine and isoleucine on which the two Committees agreed well, particularly as the scoring pattern was meant to apply for children younger than 2 years of age. The values for amino acids other than lysine are not greatly different from the NRC values. Pineda et al. (28) indicated that the INCAP tryptophan value may be an overestimate. From examination of Table 12, it would appear that the two Committees may have been in closer agreement than the proposed scoring patterns would suggest, if the recalculated values given there are accepted. If consideration is to be given by FAO/WHO to revising the amino acid scoring pattern, it would seem wise to review critically the information discussed in relation to this table.

In relation to the question of sulfur-containing amino acid requirements, the information presented by the INCAP group (28, 29) and that of Fomon et al. (29 and earlier studies) should be reviewed carefully. The analysis presented here would indicate that the FAO/WHO value for methionine plus cystine is high. The INCAP and NRC reports also support this view. Interestingly the pattern proposed by the NRC Committee resembles closely that of cow's milk (25), with which Fomon and associates (1, 2) observed satisfactory growth rates of 4.5 to 6-month old infants consuming only 1.5 gm of protein per kg of body weight per day. These observations also support the validity of a lower value for total sulfur-containing amino acids than that in the present FAO/WHO amino acid scoring pattern.

REFERENCES

1. Food and Agriculture Organization/World Health Organization (1973). WHO Technical Report Series No. 522, FAO Nutrition Meetings Report Series No. 52. FAO/WHO, Geneva.

2. Williams, H. H., Harper, A. E., Hegsted, D. M., Arroyave, G. and Holt, L. E. Jr. (1974) Nitrogen and amino acid requirements. In: Improvement of Protein Nutriture, National Academy of Sciences, Washington, D.C., pp. 23–63.

3. Rose, W. C. (1957) The amino acid requirements of adult man. Nutr. Abstr. Rev. 27: 631–647.

4. Leverton, R. M. (1959) Amino acid requirements of young adults. In: Protein and Amino Acid Nutrition (A. A. Albanese, Ed.), pp. 477–506, Academic Press, New York.

5. Reynolds, M. S., Steele, D. L., Jones, E. M. and Baumann, C. A. (1958) Nitrogen balances of women maintained on various levels of methionine and cystine. J. Nutr. 64: 99–111.

6. Rose, W. C. and Wixom, R. L. (1955) The amino acid requirements of man. XIII. The sparing effect of cystine on the methionine requirement. J. Biol. Chem. 216: 763–773.

7. Burrill, L. M. and Schuck, C. (1964) Phenylalanine requirements with different levels of tyrosine. J. Nutr. 83: 202–208.

8. Rose, W. C. and Wixom, R. L. (1955) The amino acid requirements of man. XVI. The role of the nitrogen intake. J. Biol. Chem. 217: 997–1004.

9. Hegsted, D. M. (1963) Variation in requirements of nutrients--amino acids. Fed. Proc. 22: 1424–1430.

10. Nakagawa, I., Takahashi, T. and Suzuki, T. (1960) Amino acid requirements of children. J. Nutr. 71: 176–181.

11. Nakagawa, I., Takahashi, T. and Suzuki, T. (1961) Amino acid requirements of children: Isoleucine and leucine. J. Nutr. 73: 186–190.

12. Nakagawa, I., Takahashi, T. and Suzuki, T. (1961) Amino acid requirements of children: Minimal needs of lysine and methionine based on nitrogen balance method. J. Nutr. 74: 401–407.

13. Nakagawa, I., Takahashi, T., Suzuki, T. and Kobayashi, K. (1962) Amino acid requirements of children: Minimal needs of threonine, valine and phenylalanine based on nitrogen balance method. J. Nutr. 77: 61–68.

14. Nakagawa, I., Takahashi, T., Suzuki, T. and Kobayashi, K. (1963) Amino acid requirements of children: Minimal needs of tryptophan, arginine and histidine based on nitrogen balance method. J. Nutr. 80: 305–310.

15. Nakagawa, I., Takahashi, T., Suzuki, T. and Kobayashi, K. (1964) Amino acid requirements of children: Nitrogen balance at the minimal level of essential amino acids. J. Nutr. 83: 115–118.

16. Snyderman, S. E. (1974) Heritable Disorders of Amino Acid Metabolism: Patterns of Clinical Expression and Genetic Variation (W. L. Nyhan, Ed.), pp. 641–651, John Wiley & Sons, New York.

17. Holt, L. E., Jr. and Snyderman, S. E. (1965) Protein and amino acid requirements of infants and children. Nutr. Abstr. Rev. 35: 1–13.

18. Fomon, S. J. (1961) Factors influencing retention of nitrogen by normal full-term infants. In: Progress in Meeting Protein Needs of Infants and Preschool Children. Publ. 843, National Academy of Sciences, Washington, D. C., pp. 343–353.

19. Fomon, S. J. and Filer, L. J. Jr. (1967) Amino acid requirements for normal growth. In: Amino Acid Metabolism and Genetic Variation (W. L. Nyhan, Ed.), McGraw-Hill, New York, pp. 391–402.

20. Orr, M. L. and Watt, B. K. (1968) Amino acid content of foods. Home Econ. Res. Rep. No. 4 ARS-USDA.U.S. Government Printing Office, Washington, D.C., 82 pp.

21. Fomon, S. J. (1974) Infant Nutrition, Second Edition.
W. B. Saunders Co., Philadelphia.

22. Snyderman, S. E., Holt, L. E. Jr., Dancis, J., Roitman, E., Boyer, A. and Balis, M. E. (1962) “Unessential” nitrogen: A limiting factor for human growth. J. Nutr. 78: 57–72.

23. FAO (Food and Agriculture Organization). (1957) Protein requirements. FAO Nutr. Stud. No. 16, FAO, Rome, 52 pp.

24. FAO/WHO (Food and Agriculture Organization/World Health Organization) (1965) Protein requirements. WHO Tech. Rep. Ser. No. 301; FAO Nutr. Meet. Rep. Ser. No. 37. WHO, Geneva, 71 pp.

25. Harper, A. E. (1979) Human requirements for lysine and sulfur-containing amino acids. In: Soy Protein and Human Nutrition. Academic Press, Inc., New York, pp. 171–186.

26. Harper, A. E. (1977) Amino acid requirements -- general. In: Clinical Nutrition Update - Amino Acids (H. L. Greene, M. A. Holliday and H. N. Munro, Eds.), AMA, Chicago, pp. 58–65.

27. Fomon, S. J., Ziegler, E. E., Filer, L. J. Jr., Nelson, S. E. and Edwards, B. B. (1979) Methionine fortification of a soy protein formula fed to infants. Am. J. Clin. Nutr. 32: 2460–2471.

28. Pineda, O., Torun, B., Viteri, F. E. and Arroyave, G. (1981) Protein quality in relation to estimates of essential amino acids requirements. In: Proteins for Humans: Assessment and In vitro Estimation (J. S. Adkins and D. T. Hopkins, Eds.), Avi Publishing, Westport, CT, in press.

29. Torun, B., Pineda, O., Vieteri, F. E. and Arroyave, G. (1981) Use of amino acid composition data to predict protein nutritive value for children with specific reference to new estimates of their essential amino acid requirements. In: Proteins for Humans: Assessment and In vitro Estimation (J. S. Adkins and D. T. Hopkins, Eds.), Avi Publishing, Westport, CT, in press.