Lane Molly J, McNair James N, Rediske Richard R, Briggs Shannon, Sivaganesan Mano, Haugland Richard
Annis Water Resources Institute, Grand Valley State University, Muskegon, MI 49401, USA.
Michigan Department of Environment, Great Lakes, and Energy (EGLE), 525 W. Allegan St., Lansing, MI 48909, USA.
Water (Basel). 2020 Mar 11;12(3):1-775. doi: 10.3390/w12030775.
Draft method C is a standardized method for quantifying densities in recreational waters using quantitative polymerase chain reaction (qPCR). The method includes a Microsoft Excel workbook that automatically screens for poor-quality data using a set of previously proposed acceptance criteria, generates weighted linear regression (WLR) composite standard curves, and calculates target gene copies in test samples. We compared standard curve parameter values and test sample results calculated with the WLR model to those from a Bayesian master standard curve (MSC) model using data from a previous multi-lab study. The two models' mean intercept and slope estimates from twenty labs' standard curves were within each other's 95% credible or confidence intervals for all labs. gene copy estimates of six water samples analyzed by eight labs were highly overlapping among labs when quantified with the WLR and MSC models. Finally, we compared multiple labs' 2016-2018 composite curves, comprised of data from individual curves where acceptance criteria were not used, to their corresponding composite curves with passing acceptance criteria. Composite curves developed from passing individual curves had intercept and slope 95% confidence intervals that were often narrower than without screening and an analysis of covariance test was passed more often. The Excel workbook WLR calculation and acceptance criteria will help laboratories implement draft method C for recreational water analysis in an efficient, cost-effective, and reliable manner.
方法C草案是一种使用定量聚合酶链反应（qPCR）对娱乐用水中的密度进行定量的标准化方法。该方法包括一个Microsoft Excel工作簿，它使用一组先前提出的验收标准自动筛选质量较差的数据，生成加权线性回归（WLR）复合标准曲线，并计算测试样品中的目标基因拷贝数。我们使用先前一项多实验室研究的数据，将WLR模型计算的标准曲线参数值和测试样品结果与贝叶斯主标准曲线（MSC）模型的结果进行了比较。对于所有实验室，来自20个实验室标准曲线的两个模型的平均截距和斜率估计值均在彼此的95%可信区间或置信区间内。当用WLR和MSC模型进行定量时，八个实验室分析的六个水样的基因拷贝估计值在各实验室之间高度重叠。最后，我们将多个实验室2016 - 2018年的复合曲线（由未使用验收标准的单个曲线数据组成）与其通过验收标准的相应复合曲线进行了比较。由通过的单个曲线生成的复合曲线的截距和斜率95%置信区间通常比未进行筛选时更窄，并且协方差分析测试更常通过。Excel工作簿中的WLR计算和验收标准将有助于实验室以高效、经济有效且可靠的方式实施方法C草案用于娱乐用水分析。
