Dear Kecheng,

Are you controlling protein expression?

FRET will dramatically depend on the concentration of acceptors around the donors. So, if you have cells with distinct levels of protein expression, it will eventually lead to the recovery of different efficiency values among your cell population. 

Of course, dependind on your system and the kind of proteins you are studying, it is possible that FRET is different in different cell compartments.

Hope it was helpful.

Yes.  This is possible and likely due to different FRET efficiencies in different areas of cell and cell to cell variations.  It sounds like your positive control was not as variable? In that case a good way to report the data might be to capture the variation in the numbers, so perhaps a plot of the distribution of the values, liek a histogram would be a good comparison.

I think that various FRET efficiency  in different parts of a cell is to be expected since medium viscosity and electric permittivity varies.

You can normalise your results for a particular area of interest. If for example your interaction occurs in the nucleus, then you would want to compute FRET for 30 nuclei for example, then for each nucleus measure the nuclear surface area  and divide your FRET EFF. by the nuclear surface area, then compute an average for these values. We had the same problem and found that normalisation to an area of interest worked pretty well.

http://pubs.acs.org/doi/abs/10.1021/acs.molpharmaceut.5b00478

Dear Maria, 

Thanks for your suggestion, those two protein do express in different level, nowadays, we are just normalizing to the expression level(based on the image).  We are thinking maybe FLIM-FRET is better for our purpose, since it measure the half life. Thanks for your advice. We do need always keep protein expression  in mind. 

Best, 

Kecheng

Thanks Jerilyn,  since we clone Venus-link-Cerulean as positive Ctrl, it always give robust readout. The thing is that we get various FRET efficiency in different cellular part, i think your idea( a plot of value distribution) is really nice, i am now trying figure out what this plot finally looks like, hope informative. Thanks. 

Best, 

Ken

Hi Andrzej, Thanks for your explanation of this various value. Since it is reasonable, i don't worry about results now, just need more experiment  to make it looks elegant. Thanks very much. 

Best, 

Ken