TABLE 7

Variance component analyses (% of the variability by source)

Cell Line Data OnlyAll Systems
ResponseA>B w/o InhibitorB>A w/o InhibitorLog ER1 w/o InhibitorA>B 
with InhibitorB>A 
with InhibitorLog ERa with InhibitorLog IC50 A>BaLog IC50 B>AaLog IC50 B>A 
or Uptake into 
Inside-Out Vesiclea
Cells/systems908780131428
Labs778262787945353529
Cell/system by experimentb000000Not modeledNot modeledNot modeled
Lab by experimentc914206829Not modeledNot modeledNot modeled
Error5410852552d51 d43 d
Total100100100100100100100100100
  • a Log ER and log IC50 were used in variance component analysis since data are log normally distributed. ER, efflux ratio.

  • b Interaction between experiment and cell/system (for each time the experiment was performed, this is the variability in probe substrate transport that can be ascribed to differences between the three cell lines/four experimental systems).

  • c Interaction between experiment and laboratory (for each time the experiment was performed, this is the variability in probe substrate transport that can be ascribed to differences between the laboratories).

  • d Variability due to replication, which in this case constitutes interactions between experiment and cell/system, experiment and laboratory, well-to-well variability and measurement variability that could not be further deconvoluted because only a single IC50 value for each inhibitor was used. While the vesicle laboratories generated two sets of IC50 values for most compounds, only day 1 vesicle-IC50 values were included for consistency with the cell-based systems.