Hi
Is it acceptable to adjust results (RFU) cell viability assay (ATP), According to number of cells seeded, I have done three replicates of an experiment were the number of cells I have seeded in one deplicate different from the other two, by which it is very clear in the results - so I adjusted the results according to the ratio between the cells seeded (Caco-2 cells)?
Well… You could. It's certainly more correct than presenting data that you know is different not because of the experimental conditions but due to a huge confounding error. But it’s not great science.
Cells that are more confluent tend to be happier than cells that are sparsely seeded. If you have cells that are 25%, 50%, 75%, and 100% and you treat them with something that they don't like you will find that in the majority of instances the impact on the less confluent cells is more severe. You have two different populations of cells. One is more confluent, there are more cell-cell adhesions, more cytokines and other factors being released into the media, depending on the cell type more ECM is being laid down etc. The other has less of all these things that are known to have an effect on cell behaviour.
So even if you correct for the cell number it doesn't change the fact that you aren't comparing the same population of cells. The best you could hope for is that the effect of the difference in cell number was less than the effect of your treatment and you could argue that the data is probably, maybe, mostly, kind of right. But this is very bad practice since in the worst-case scenario is that the effect of cell number dwarfs any changes you were expecting to see with your treatment groups to such a significant extent you may as well not have treated them at all.
If you're using Caco-2 cells. A cell line which is known to undergo multiple changes after reaching confluence, and you want to do good science. No, you can't. Caco-2 cells are a good example of a cell line where there can be a really big impact from having different numbers of cells. It's never good science to be treating two populations that you suspect to be different as if they are the same. But if you’re looking at Caco-2 cells you're probably looking at a worst-case scenario.
Even if you have to plate out 100 wells of cells to get 10 that are the same, you'll find in the long run you'll be much better off and be forever thankful towards your past self that you did it until it's done right, rather than take shortcuts, and for what? An extra hour or two of free time? You may well end up spending triple, even quintuple that pouring over largely nonsensical data scrounging for some kind of significance. That is a very slippery slope to be going down.
- Badges
- Science method