M R Facer | PubMed Health

PubMed1998

Statistical models for quantitative bioassay.

We discuss various statistical approaches useful in the analysis of nutritional dose-response data with a continuous response. The emphasis is on the multivariate case with several predictors. The methods which will be discussed can be classified into parametric models, including change-point models, and nonparametric models, which rely on smoothing methods such as weighted local linear fitting. The methods will be illustrated with the analysis of data generated from a folate depletion-repletion bioassay experiment conducted on rats, where the measured growth rate of the rate is the response variable. We also discuss the biological conclusions that can be drawn from applying various statistical methods to this data set.

Animal Nutritional Physiological Phenomena↗

PubMed1996

Statistical interaction model for exchangeability of food folates in rat growth bioassay.

The comparative value of several sources of dietary folate in promoting growth of folate-depleted rats was determined in a folate depletion-repletion rat growth bioassay. Folate-depleted rats were fed an amino acid-based diet supplemented with 11 different concentrations of folate (227, 272, 317, 363, 408, 454, 499, 544, 590, 635 and 680 nmol/kg) from each of 12 different sources of folate (folic acid, fried beef liver, cooked pinto beans individually, or as 1/3, 1/1, or 3/1 combinations of folate from the folic acid/beans, folic acid/beef liver and beans/beef liver) for a total of 132 treatments. Growth response to folic acid and bean folate was linear, whereas that to beef liver folate was distinctly nonlinear, beef liver folate being more potent at lower dietary concentrations but less potent at higher concentrations compared with folic acid and bean folate. Folic acid and bean folate were equivalent to and exchangeable with one another in promoting growth. Beef liver folate and folic acid/bean folate had an interactive effect in promoting growth. The nature of the interaction was antagonistic in that the presence of folic acid and/or bean folate reduced the efficacy of beef liver folate and vice versa. Beef liver folate is not exchangeable with either folic acid or bean folate. We conclude that food folates generally are not exchangeable and do interact adversely. A statistical interaction model that predicted the growth-promoting effect of several sources of dietary folate was developed and validated.

Animals↗