Metagenomic next generation sequencing (mNGS) is increasingly recognized as an important complementary tool to targeted human and animal infectious disease diagnostics. It is, however, sensitive to biases and errors that are currently not systematically evaluated by the implementation of quality controls (QC) for the diagnostic use of mNGS. We evaluated a commercial reagent (Mengovirus extraction control kit, CeraamTools, bioMérieux) as an exogenous internal control for mNGS. It validates the integrity of reagents and workflow, the efficient isolation of viral nucleic acids and the absence of inhibitors in individual samples (verified using a specific qRT-PCR). Moreover, it validates the efficient generation of viral sequence data in individual samples (verified by normalized mengoviral read counts in the metagenomic analysis). We show that when using a completely random metagenomics workflow: (1) Mengovirus RNA can be reproducibly detected in different animal sample types (swine feces and sera, wild bird cloacal swabs), except for tissue samples (swine lung); (2) the Mengovirus control kit does not contain other contaminating viruses that may affect metagenomic experiments (using a cutoff of minimum 1 Kraken classified read per million (RPM)); (3) the addition of 2.17 × 10⁶ Mengovirus copies/mL of sample does not affect the virome composition of pig fecal samples or wild bird cloacal swab samples; (4) Mengovirus Cq values (using as cutoff the upper limit of the 99 % confidence interval of Cq values for a given sample matrix) allow the identification of samples with poor viral RNA extraction or high inhibitor load; (5) Mengovirus normalized read counts (cutoff RPM > 1) allow the identification of samples where the viral sequences are outcompeted by host or bacterial target sequences in the random metagenomic workflow. The implementation of two QC testing points, a first one after RNA extraction (Mengoviral qRT-PCR) and a second one after metagenomic data analysis provide valuable information for the validation of individual samples and results. Their implementation in addition to external controls validating runs or experiments should be carefully considered for a given sample type and workflow.

新一代元基因组测序（mNGS）被越来越多地视为针对人类和动物传染病诊断的重要补充工具。但是，它对当前尚未通过实施质量控制（QC）进行mNGS诊断的系统性评估的偏见和错误敏感。我们评估了一种商业试剂（Mengovirus提取控制套件（CeraamTools，bioMérieux）作为mNGS的外源内部对照。它验证了试剂和工作流程的完整性，有效分离病毒核酸以及单个样品中不存在抑制剂的情况（已使用特定的qRT-PCR进行了验证）。此外，它验证了单个样品中病毒序列数据的高效生成（通过宏基因组学分析中的标准化芒果病毒读取计数进行验证）。我们表明，当使用完全随机的宏基因组学工作流程时：（1）除组织样品（猪肺）外，在不同的动物样品类型（猪粪便和血清，野禽泄殖腔拭子）中都可以可复制地检测到Mengovirus RNA；（2）芒果病毒对照试剂盒不包含其他可能影响宏基因组学实验的污染病毒（使用百万分之1的Kraken分类读值（RPM）的下限）；（3）添加2.17×10 ^6个 Mengovirus拷贝/ mL样品不会影响猪粪便样品或野鸟泄殖腔拭子样品的病毒组成；（4）Mengovirus Cq值（使用给定样品基质的Cq值的99％置信区间的上限作为临界值）可以鉴定病毒RNA提取不良或抑制剂载量高的样品；（5）芒果病毒归一化的读取计数（截止RPM> 1）允许在随机宏基因组学工作流程中鉴定病毒序列被宿主或细菌靶序列竞争的样本。实施两个QC测试点，第一个在RNA提取后进行（Mengoviral qRT-PCR），第二个在宏基因组数据分析后，为验证单个样品和结果提供了有价值的信息。对于给定的样品类型和工作流程，应仔细考虑除验证运行或实验的外部控制外，还应考虑其实施。