Unveiling Errors in Soil Microbial Community Sequencing: A Case for Reference Soils and Improved Diagnostics for Nanopore Sequencing

Sequencing platform and workflow strongly influence microbial community analyses through potential errors at each step. Effective diagnostics and experimental controls are needed to validate data and improve reproducibility. This cross-laboratory study evaluates sources of variability and error at three main steps of a standardized amplicon sequencing workflow (DNA extraction, polymerase chain reaction [PCR], and sequencing) using Oxford Nanopore MinION to analyze agricultural soils and a simple mock community. Variability in sequence results occurs at each step in the workflow with PCR errors and differences in library size greatly influencing diversity estimates. Common bioinformatic diagnostics and the mock community are ineffective at detecting PCR abnormalities. This work outlines several diagnostic checks and techniques to account for sequencing depth and ensure accuracy and reproducibility in soil community analyses. These diagnostics and inclusion of a reference soil can help ensure data validity and facilitate comparison of multiple sequencing runs within and between laboratories.

测序平台与测序流程在每一步均可能引入误差，因此会对微生物群落分析产生显著影响。为验证数据质量并提升实验可重复性，亟需采用有效的诊断分析手段与实验对照体系。本项跨实验室研究针对标准化扩增子测序（amplicon sequencing）流程的三大核心环节——DNA提取、聚合酶链式反应（PCR）以及测序环节本身——展开变异与误差来源解析，实验采用牛津纳米孔MinION（Oxford Nanopore MinION）测序平台，对农业土壤样本与简易模拟群落（mock community）进行测序分析。测序结果的变异贯穿流程全环节，其中PCR误差与文库（library）大小差异会对群落多样性评估结果造成显著影响。常规生物信息学诊断方法与模拟群落均无法有效检出PCR异常状况。本研究提出了若干可用于校正测序深度、保障土壤微生物群落分析准确性与可重复性的诊断检测方案与技术策略。上述诊断方法结合参考土壤样本的纳入，可有效保障数据有效性，并推动实验室内部及跨实验室的多轮测序结果比对工作。