Effect of Quantitative Polymerase Chain Reaction Data Analysis Using Sample Amplification Efficiency on Microbial Source Tracking Assay Performance and Source Attribution
Effect of Quantitative Polymerase Chain Reaction Data Analysis Using Sample Amplification Efficiency on Microbial Source Tracking Assay Performance and Source Attribution
复制标题
DOI：
10.1021/acs.est.0c01559
复制
发表时间：
2020-07-07
影响因子：
11.4
通讯作者：
Sirikanchana, Kwanrawee
中科院分区：
文献类型：
--
作者：
Kongprajug, Akechai;Chyerochana, Natcha;Sirikanchana, Kwanrawee
广泛使用的微生物源追踪（MST）技术，定量聚合酶链反应（qPCR），量化污染水中的宿主特异性基因丰度，以识别和优先考虑污染源。本研究使用样本PCR效率（LinRegPCR模型）表征qPCR数据分析对基因丰度的影响，并将其与基于标准曲线的方法（混合模型）进行比较。评价了5种qPCR测定法：通用GenBac 3、人特异性HF 183/BFDrev和CPQ_056、猪特异性Pig-2-Bac和牛特异性Bac 3qPCR测定法。LinRegPCR模型增加了低拷贝扩增，尤其是在HF 183/BFDrev测定中，从而将特异性降低至0.34。通过LinRegPCR方法，观察到混合粪便和污水样品（n = 147）的差异高达1.41 log（10）拷贝/g和0.41 log（10）拷贝/100 mL，分别对应于18.2%的增加和6.4%的减少。淡水样本（n = 48）显示两种模型之间的最大差异为1.95 log（10）拷贝/100 mL。两种模型在54.55%的环境样本中显示出相同的来源，同时，LinRegPCR方法改善了混合模型在29.55%的样本中无法识别来源的情况。本研究强调了qPCR MST检测需要标准化数据分析方案，以确保实验室间的一致性和可比性。
The widely used microbial source tracking (MST) technique, quantitative polymerase chain reaction (qPCR), quantifies host-specific gene abundance in polluted water to identify and prioritize contamination sources. This study characterized the effects of a qPCR data analysis using the sample PCR efficiencies (the LinRegPCR model) on gene abundance and compared them with the standard curve-based method (the mixed model). Five qPCR assays were evaluated: the universal GenBac3, human-specific HF183/BFDrev and CPQ_056, swine-specific Pig-2-Bac, and cattle-specific Bac3qPCR assays. The LinRegPCR model increased the low-copy amplification, especially in the HF183/BFDrev assay, thus lowering the specificity to 0.34. Up to 1.41 log(10) copies/g and 0.41 log(10) copies/100 mL differences were observed for composite fecal and sewage samples (n = 147) by the LinRegPCR approach, corresponding to an 18.2% increase and 6.4% decrease, respectively. Freshwater samples (n = 48) demonstrated a maximum of 1.95 log(10) copies/100 mL difference between the two models. Identical attributing sources by both models were shown in 54.55% of environmental samples; meanwhile, the LinRegPCR approach improved the inability to identify sources by the mixed model in 29.55% of the samples. This study emphasizes the need for a standardized data analysis protocol for qPCR MST assays for interlaboratory consistency and comparability.