van Dam N M, Poppy G M
Netherlands Institute of Ecology (NIOO-KNAW), Multitrophic Interactions Department, P.O. Box 40, 6666 ZG Heteren, The Netherlands.
Plant Biol (Stuttg). 2008 Jan;10(1):29-37. doi: 10.1055/s-2007-964961.
Plant volatile analysis may be the oldest form of what now is called plant "metabolomic" analysis. A wide array of volatile organic compounds (VOCs), such as alkanes, alcohols, isoprenoids, and esters, can be collected simultaneously from the plant headspace, either within the laboratory or in the field. Increasingly faster and more sensitive analysis techniques allow detection of an ever-growing number of compounds in decreasing concentrations. However, the myriads of data becoming available from such experiments do not automatically increase our ecological and evolutionary understanding of the roles these VOCs play in plant-insect interactions. Herbivores and parasitoids responding to changes in VOC emissions are able to perceive minute changes within a complex VOC background. Plants modified in genes involved in VOC synthesis may be valuable for the evaluation of changes in plant-animal interactions compared to tests with synthetic compounds, as they allow changes to be made within the context of a more complex profile. We argue that bioinformatics is an essential tool to integrate statistical analysis of plant VOC profiles with insect behavioural data. The implementation of statistical techniques such as multivariate analysis (MVA) and meta-analysis is of the utmost importance to interpreting changes in plant VOC mixtures. MVA focuses on differences in volatile patterns rather than in single compounds. Therefore, it more closely resembles the information processing in insects that base their behavioural decisions on differences in VOC profiles between plants. Meta-analysis of different datasets will reveal general patterns pertaining to the ecological role of VOC in plant-insect interactions. Successful implementation of bioinformatics in VOC research also includes the development of MVA that integrate time-resolved chemical and behavioural analyses, as well as databases that link plant VOCs to their effects on insects.
植物挥发性成分分析可能是现在所谓植物“代谢组学”分析的最古老形式。各种各样的挥发性有机化合物（VOCs），如烷烃、醇类、类异戊二烯和酯类，可以在实验室或野外从植物顶空同时收集。越来越快且更灵敏的分析技术能够检测出浓度越来越低的越来越多的化合物。然而，从这类实验中获得的大量数据并不会自动增加我们对这些VOCs在植物 - 昆虫相互作用中所起作用的生态和进化理解。食草动物和寄生蜂对VOC排放变化的反应能够在复杂的VOC背景中感知微小变化。与使用合成化合物的测试相比，在参与VOC合成的基因中经过修饰的植物对于评估植物 - 动物相互作用的变化可能很有价值，因为它们允许在更复杂的概况背景下进行变化。我们认为生物信息学是将植物VOC概况的统计分析与昆虫行为数据整合的重要工具。多变量分析（MVA）和荟萃分析等统计技术的应用对于解释植物VOC混合物的变化至关重要。MVA关注挥发性模式的差异而非单一化合物的差异。因此，它更类似于昆虫基于植物间VOC概况差异做出行为决策的信息处理方式。对不同数据集的荟萃分析将揭示与VOC在植物 - 昆虫相互作用中的生态作用相关的一般模式。在VOC研究中成功实施生物信息学还包括开发整合时间分辨化学和行为分析的MVA，以及将植物VOC与其对昆虫的影响联系起来的数据库。
