Skip to main content
ARS Home » Plains Area » Grand Forks, North Dakota » Grand Forks Human Nutrition Research Center » Dietary Prevention of Obesity-related Disease Research » Research » Publications at this Location » Publication #97444
Title: PREDICTION OF DIETARY COPPER, ZINC AND IRON BASED ON RESPONSE FACTORS IN BLOOD AND PLASMA OF RATS

Author
item Zaslavsky, Boris
item Uthus, Eric

Submitted to: Experimental Biology
Publication Type: Abstract Only
Publication Acceptance Date: 4/17/1999
Publication Date: N/A
Citation: N/A

Interpretive Summary:

Technical Abstract: In a 4x4x3 factorial experiment, blood and plasma parameters were measured in rats (N=6/group) fed a semi-purified diet with graded concentrations of Cu, Zn and Fe in all combinations: Cu at 0.75, 1.5, 3 or 6 ug/g, Zn at 5, 10, 20 or 40 ug/g, and Fe at 12, 24 or 48 ug/g. To reveal the predominant response factors, 16 blood and 3 plasma parameters were subjected to principal component analysis, factor analysis, and discriminant analysis. These 3 methods were tested for their ability to predict dietary groups by using only the blood or plasma data under the ad hoc assumption that the diet of a particular rat is unknown. Our approach is motivated by the aim to develop algorithms to calculate individual RDAs or DRIs on the basis of inexpensive and nondestructive biological analyses. Based on blood data and by using principal component analysis, dietary Fe could be predicted (100%) for Fe=12 ug/g and Fe>12 ug/g. Use of logistic regression following principal component analysis resulted in the ability to distinguish all 3 Fe groups (79.3%). Discriminant analysis, which linearly combines 16 blood characteristics into one characteristic, resulted in 92.1% success. We were unable to distinguish the Cu and Zn groups with the 16 blood parameters by any of the methods. However, when only the plasma parameters (ceruloplasmin, cholesterol and glucose) were used, we were able to distinguish Cu intake. Factor analysis alone was unable to distinguish Cu intake. However, logistic analysis by the first factor of factor analysis and discriminant analysis were able to distinguish the 4 dietary Cu concentrations with 83.2% and 87% success, respectively. When using only the blood/plasma data, dietary Fe and Cu can be predicted sufficiently well, while Zn cannot.