DIRECTIONAL SELECTION IN TEMPORALLY REPLICATED STUDIES IS REMARKABLY CONSISTENT

DIRECTIONAL SELECTION IN TEMPORALLY REPLICATED STUDIES IS REMARKABLY CONSISTENT
复制标题

DOI:
10.1111/j.1558-5646.2011.01444.x
复制
发表时间:
2012-02-01
期刊:
影响因子:
3.3
通讯作者:
Hadfield, Jarrod D.
Hadfield, Jarrod D.
中科院分区:
环境科学与生态学2区
文献类型:
--
作者:
Morrissey, Michael B.;Hadfield, Jarrod D.

文献摘要

被引文献

相似文献

选择的时间变化是进化结果的基本决定因素。最近的一篇论文对自然种群选择的时间变化进行了综合分析。作者得出的结论是,随着时间的推移,选择的强度和方向存在很大的变化,但承认抽样误差会导致选择的估计值比真实值变化更大。我们使用技术重新分析他们的数据集,这些技术考虑了采样误差对扩大明显变化水平的必要影响,并表明方向选择随着时间的推移在幅度和方向上都非常恒定。因此,我们不能声称现有数据支持选择中存在显着的时间异质性。尽管如此,我们推测选择的时间变化可能很重要,但有充分的理由解释为什么它可能不会出现在可用数据中。这些新的分析强调了应用估计选择分布参数的技术的重要性,而不是估计选择分布的参数(这将反映抽样误差和选择的实际变化);事实上,尽管有针对前者的方法,但在对自然选择方面的综合评价中,对后者的关注却很常见,并且可能导致严重的误解。
Temporal variation in selection is a fundamental determinant of evolutionary outcomes. A recent paper presented a synthetic analysis of temporal variation in selection in natural populations. The authors concluded that there is substantial variation in the strength and direction of selection over time, but acknowledged that sampling error would result in estimates of selection that were more variable than the true values. We reanalyze their dataset using techniques that account for the necessary effect of sampling error to inflate apparent levels of variation and show that directional selection is remarkably constant over time, both in magnitude and direction. Thus we cannot claim that the available data support the existence of substantial temporal heterogeneity in selection. Nonetheless, we conject that temporal variation in selection could be important, but that there are good reasons why it may not appear in the available data. These new analyses highlight the importance of applying techniques that estimate parameters of the distribution of selection, rather than parameters of the distribution of estimated selection (which will reflect both sampling error and real variation in selection); indeed, despite availability of methods for the former, focus on the latter has been common in synthetic reviews of the aspects of selection in nature, and can lead to serious misinterpretations.