Choice of an estimate of genetic variance from twin data.
Choice of an estimate of genetic variance from twin data.
复制标题
从双胞胎数据中选择遗传方差的估计值。
DOI：
--
复制
发表时间：
1974
影响因子：
9.8
通讯作者：
J. Norton
中科院分区：
文献类型：
--
作者：
J. Christian;K. Kang;J. Norton
致编辑：在一篇题为“从双胞胎数据中选择遗传方差估计”的文章中，Christian等人主张使用同卵（MZ）和异卵（DZ）双胞胎的配对间和配对内均方来获得双胞胎遗传方差的估计值。每当总均方差异超出偶然预期时。他们建议使用0.2的水平作为使用遗传方差的组合估计的标准[ 1 ]。这种方法的有效性关键取决于隐藏的假设，即MZ和DZ的总遗传方差相等。在忽略显性、上位性、环境效应和遗传环境互作的情况下，同卵双胞胎的双对均方的加性方差（或2AMz）与异卵双胞胎的双对均方的加性方差（0-2ADZ）的比值可能大于、等于或小于1，这取决于202 AMZ是大于、等于还是小于3/2 o-2ADz。因此，在没有总遗传方差相等的先验知识的情况下，双对均方对于遗传方差的估计似乎没有什么价值。请注意，即使总遗传方差不相等，DZ与MZ双胞胎的配对内方差之比仍然是DZ双胞胎中存在遗传效应的有效检验。这是因为MZ内均方不包括遗传变异。在检测到遗传变异的证据后，研究者必须判断MZ和DZ双胞胎的配对内环境方差是否真正具有可比性，该性状是否可能受到遗传-环境相互作用或“双胞胎情况”的影响[2]，以及DZ双胞胎的遗传方差估计值是否与单胞胎有任何相关性。在解释配对内方差时，特别值得关注的是MZ和DZ孪生生物学差异所产生的潜在偏倚。已知存在胎盘形成的重要差异[3]，这可能导致MZ双胞胎的变异[4]。在DZ双胞胎中，已经在出生体重[5]和酶水平[6]的性别不同的双胞胎中记录了协同作用，这提高了协同作用可能是一种更普遍的现象的可能性，也影响了遗传上不同的性别相同的DZ双胞胎。最后，如果人类中的超胎发生频率相当高[7，8]，这将构成经典双胞胎研究的主要偏倚。然而，这些外来变异的潜在来源中的大多数（如果不是全部的话）都会影响配对内和配对间的成分。由于这个原因，在配对内比较因生物学原因被拒绝的情况下，似乎不太可能从配对内比较中挽救出一个有效的遗传效应检验。所涉及的问题很容易通过一个假设的例子来说明。如果
To the Editor: In an article entitled "Choice of an Estimate of Genetic Variance from Twin Data," Christian et al. advocate the use of both the among-pair as well as the within-pair mean squares of monozygotic (MZ) and dizygotic (DZ) twins to obtain estimates of the twin genetic variance whenever the total mean squares differ more than could be expected by chance. They recommend an a level of 0.2 as a criterion for using their combined estimate of genetic variance [ 1 ]. The validity of this approach depends critically upon the hidden assumption that the total genetic variances of MZ and DZ are equal. Ignoring dominance, epistasis, environmental effects and genetic environmental interactions, the ratio of the additive variance in the among-pair mean square of MZ twins (or2AMz) to the additive variance in the among-pair mean square of DZ twins (0-2ADZ) could be greater than, equal to, or less than 1 depending upon whether 202AMZ is greater than, equal to, or less than 3/2o-2ADz. Thus, in the absence of prior knowledge that the total genetic variances are equal, the among-pair mean squares would appear to be of little value for the estimation of genetic variance. Note that even if the total genetic variances are not equal, the ratio of the within-pair variance of DZ to MZ twins remains a valid test for the presence of a genetic effect in DZ twins. This is because the within MZ mean square does not include genetic variation. Having detected evidence for genetic variation, the investigator must then judge whether the within-pair environmental variances of MZ and DZ twins are truly comparable, whether the trait is one which may be influenced by genetic-environmental interactions or the "twin situation" [2], and whether an estimate of genetic variance in DZ twins has any relevance to singletons. Of particular concern in the interpretation of within-pair variances are potential biases arising from differences in the biology of MZ and DZ twinning. Important differences in placentation are known to exist [3] which can contribute to variation in MZ twins [4]. In DZ twins the synergistic interactions that have been documented in unlike-sexed pairs for birth weight [5] and enzyme levels [6] raise the possibility that synergism may be a more general phenomenon, affecting genetically dissimilar like-sexed DZ twins as well. Finally, if superfetation occurs with appreciable frequency in man [7, 8], it would constitute a major bias in classical twin studies. However, most if not all of these potential sources of extraneous variation would influence the among as well as the within-pair components. For this reason, it seems unlikely that a valid test for genetic effects could be salvaged from an among-pair comparison in situations where the within-pair comparison has been rejected on biologic grounds. The issues involved are readily illustrated by a hypothetical example. If the