Hi,

I was interested in the cytokine secretion of various conditions and I conducted N=4 cytokine arrays to compare these conditions. (ref: R&D systems ARY022B)

However, in each experiment, the background that I substracted to the signal did not seem to be enough for a normalization. I still see some different general levels of signal from one experiment to the other. In each experiment, the comparison between each condition is kind of similar, but if I just mean all the replicates signal, I end up with one experiment over-represented compared to the others.

To answer that, I computed Z-scores for each experiments and meaned them over the replicates. What do you think of this method? I couldn't find in the litterature a standardized and proper way to treat this kind of data.

In the end, I feel like presenting the data as a heatmap, is it suitable to show these 'mean z-score' in a heatmap?

Bests,

Theo