| Abstrakt: |
Previous research demonstrated the existence of a major gene (hg), expressed as a homozygous recessive, that increases postweaning growth 60% in C57B1/6 mice (line Ch) compared to the same genetic stock without the major gene (line CH). Effects of the hg gene on the partial energetic efficiencies of protein and fat deposition in mice between 21 and 42 d of age were examined by using comparative slaughter data and multiple regression techniques. Efficiency estimates for fat deposition were biologically impossible and unstable, probably as a result of high correlations between the independent variables (multicollinearity). Use of biased regression techniques gave rise to biologically possible estimates for efficiencies of fat and protein gain. However, biased regression procedures introduced subjectivity in an otherwise objective analysis. Since these regression techniques produce biologically possible estimates, the physiological implications of the results are discussed in relation to protein and fat turnover. Although the usefulness of these statistical methods for predictive purposes is not questioned, it is concluded that they do little toward advancing our understanding of energy metabolism. Rather, we suggest that mechanistic models be used wherein specific biological processes are presented separately and effects of energy intake on these processes accommodated explicitly. |
