Data from: New environmental metabarcodes for analysing soil DNA: potential for studying past and present ecosystems

Metabarcoding approaches use total and typically degraded DNA from environmental samples to analyse biotic assemblages and can potentially be carried out for any kinds of organisms in an ecosystem. These analyses rely on specific markers, here called metabarcodes, which should be optimized for taxonomic resolution, minimal bias in amplification of the target organism group and short sequence length. Using bioinformatic tools, we developed metabarcodes for several groups of organisms: fungi, bryophytes, enchytraeids, beetles and birds. The ability of these metabarcodes to amplify the target groups was systematically evaluated by (1) in silico PCRs using all standard sequences in the EMBL public database as templates, (2) in vitro PCRs of DNA extracts from surface soil samples from a site in Varanger, northern Norway, and (3) in vitro PCRs of DNA extracts from permanently frozen sediment samples of late-Pleistocene age (~ 16 000–50 000 yr BP) from two Siberian sites, Duvanny Yar and Main River. Comparison of the results from the in silico PCR with those obtained in vitro showed that the in silico approach offered a reliable estimate of the suitability of a marker. All target groups were detected in the environmental DNA, but we found large variation in the level of detection among the groups and between modern and ancient samples. Success rates for the Pleistocene samples were highest for fungal DNA, whereas bryophyte, beetle and bird sequences could also be retrieved, but to a much lesser degree. The metabarcoding approach has considerable potential for biodiversity screening of modern samples and also as a paleoecological tool.

元条形码测序（metabarcoding）技术利用环境样本中的总DNA（通常为降解态DNA）分析生物群落，理论上可用于检测生态系统中的任意生物类群。这类分析依赖于特定的标记序列——本文称之为元条形码（metabarcodes），其设计需优化以满足三项核心要求：具备足够的分类学分辨率、对目标类群的扩增偏倚极低，且序列长度较短。我们借助生物信息学工具，为真菌、苔藓植物、线蚓类、甲虫及鸟类这几类生物开发了对应的元条形码。随后通过三项实验系统评估了这些元条形码对目标类群的扩增能力：（1）以欧洲分子生物学实验室（European Molecular Biology Laboratory, EMBL）公共数据库中的全部标准序列作为模板，开展计算机模拟PCR（in silico PCR）；（2）以挪威北部瓦朗厄尔地区某采样点的表层土壤DNA提取物为模板，开展体外PCR；（3）以两处西伯利亚采样点——杜万尼亚尔（Duvanny Yar）与主河（Main River）的晚更新世（约16000~50000 yr BP）永久冻土沉积物DNA提取物为模板，开展体外PCR。将计算机模拟PCR的结果与体外实验结果对比后发现，计算机模拟方法可可靠地评估标记序列的适用性。所有目标类群均在环境DNA样本中被检出，但不同类群间以及现代样本与古样本间的检出水平存在显著差异。晚更新世样本中，真菌DNA的检出成功率最高；苔藓植物、甲虫与鸟类的序列虽也可被检出，但检出效率远低于真菌。综上，元条形码测序技术在现代样本的生物多样性筛查中具备巨大应用潜力，同时也可作为古生态学研究工具。