Durrieu G, Letellier T, Antoch J, Deshouillers J M, Malgat M, Mazat J P
Laboratoire de Mathématiques Stochastiques-Université Bordeaux II, France.
Mol Cell Biochem. 1997 Sep;174(1-2):149-56.
The mitochondrial pathologies are a heterogeneous group of metabolic disorders that are characterized by anomalies of oxidative phosphorylation, especially in the respiratory chain. The diagnosis of these pathologies involves many investigations among which biochemical study is at present the main tool. However, the analysis of the results obtained during such study remains complex and often does not make it possible to conclude clearly if a patient is affected or not by a biochemical and/or bioenergetic deficiency. This arises from two main problems: 1. The determination of control values from the whole set of variable values (affected and unaffected people). 2. The small size of the population studied and the large number of variables collected which present a rather large variability. To cope with these problems, the principal component analysis method is applied to the results obtained during our biochemical studies. This analysis makes it possible for each respiratory chain complex, to distinguish clearly two subsets of the whole population (affected and unaffected people) as well as to detect the variables which are the most discriminative.
线粒体疾病是一组异质性的代谢紊乱疾病,其特征是氧化磷酸化异常,尤其是呼吸链中的异常。这些疾病的诊断涉及多项检查,其中生化研究目前是主要手段。然而,对该研究过程中获得的结果进行分析仍然很复杂,而且往往无法明确判断患者是否存在生化和/或生物能量缺乏。这主要源于两个问题:1. 从整个变量值集合(患病和未患病个体)中确定对照值。2. 所研究人群规模较小,而收集的变量数量众多且具有相当大的变异性。为了解决这些问题,我们将主成分分析方法应用于生化研究中获得的结果。这种分析能够针对每个呼吸链复合体,清晰地区分整个人群中的两个子集(患病和未患病个体),并检测出最具区分性的变量。