Statistical uncertainty of predicted and observed effect concentrations for mixtures (ERLUX).

CA, Concentration Addition; IA, Independent Action; CI, Confidence Interval. All predictions statistically significant to the observed ECs are shown in bold. *Effect mixture concentration for effect levels higher than the lowest estimated compound maximal model asymptote are extrapolated either (i) by assuming no contribution of this compound to the overall mixture effect (toxic unit equals zero), or (ii) by setting the compounds’ toxic unit to a fixed level equalling the value at the mixture concentration producing an effect of 0.7*θmax (see Table 2). The right side of the interval corresponds to (i) and the left side to (ii), defining the range of possible CA predictions.

CA为浓度加和（Concentration Addition），IA为独立作用（Independent Action），CI为置信区间（Confidence Interval）。所有与观测到的有效浓度（ECs）具有统计学显著性的预测结果均以粗体标注。* 当效应水平高于最低估算的化合物模型最大渐近效应值时，混合物效应浓度的外推采用以下两种方式之一：(i) 假设该化合物对整体混合物效应无任何贡献（毒性单位为零）；或(ii) 将该化合物的毒性单位固定为对应产生0.7×θmax效应的混合物浓度下的取值（详见表2）。该区间的右侧对应方法(i)，左侧对应方法(ii)，由此界定了浓度加和法（CA）预测结果的可能范围。