Protocol Online logo
Top : Forum Archives: : Bioinformatics and Biostatistics

Another statistics data analysis Q - How to evaluate colony formation from sorted CHO cell (Feb/09/2007 )

Hi everyone,

I would like to ask for advice in regards to the evaluation of some FACS sorting I've done.

Basically I've treated my cells and then cell sorted according to EGFP status (cells will only be positive if the treatments worked - this will be less than 1% of the population). I sorted either 1 cell/well or 10 cells/well into 96 well plates and then looked at how many colonies formed. As a viability control I did some plates from the negative sorted population.

I then counted the colonies and determined percentage compared to the number of cells I originally sorted (for eample 14 colonies out of 150 initially sorted cells) and normalized by the percentage of colony formation in the negatively sorted group.

So now I want to compare the three treatments, but am not entirely sure how I should do that. I think I should do something like a power calculation, but I have no idea how that would work...

As an alternative I guess I could determine the number of colonies per 100 initially sorted cells and then use that number to compare between treatments, but that way I'd loose some of the data as for one treatment I sorted nearly 1000 cells whereas for the other only 100 (ran out of cells for the sort).

Any help would be most welcome

Best regards



Hi LeserattePD,

not so sure, if I understood you're question. You got three treatment groups and for every group you have determined normalized values in percent? Why not using a simple one-way ANOVA + post-hoc test to compare the groups? A power calculation won't help, as you do not get any information on the differences between the groups.
But maybe I am too simple minded and didn't get the question huh.gif

Dum Di Dum Di Dum ... KEKSE!!!