For example, I have 4 cell treatments and technical repeat wells of each with confluence data every hour for 48 hours.
Would Two-way ANOVA be a good way to observe the differences between the cell treatments over time? I want to see that the different drug concentrations effect cell proliferation overtime.
I think the aim of the experiment is
a) to estimate the rate of cell growth (over 48h) within the given tretments and
b) to estimate the differnces between these growth rates.
Finding an optimal way to analyze such data can be tricky. You may have a logistic growth curve (growth will be exponential in early time points but stopping at confluency. Then it may be best to estimate the time "needed to reach 50% confulency" for the infividual wells. Then use a linear mixed model for these values with "treatment" as fixed factor and "experiment" (technical replicates) as random factor.
These two steps might be combined in some generalized nonlinear mixed model, but I am off there...
If the growth curves are nicelly exponential (and not logistic) than it may be better to estimate the growth rates directly from the slope of a straight-line model of the log cell density versus time. Here again, differences between treatments may be analysed in a second step (as describes above) or in a mixed model with well (within experiment) and experiment as random factors.
I agree with Jochen. You can do a linear mixed model with treatment and as fixed factor and experiment as random factor. This is easily done in R with the lmer (package lme4):
model
- Badges
- Science method
- Similar topics
- Biological Science
- Histology
- Staining