View Single Post
Old 04-10-2014, 06:40 AM   #4
mbblack
Senior Member
 
Location: Research Triangle Park, NC

Join Date: Aug 2009
Posts: 245
Default

I think areyes was trying to get at two points you need to clarify.

1. Differential expression is just that, a purely relative measure of expression changes when compared to some arbitrary baseline. So you need a common baseline in order to compute differential expression across your 9 species for comparison. What, exactly, is your intended baseline? Is there an accepted common ancestral species you can use as a baseline for each of these, or a single most common species found in the same habitat? Bottom line is you need one common genome to use as the reference genome for the differential expression computation for the other 9 species you wish to compare.

E.G. In an experiment of some xenobiotic exposure to the these animals, your common baseline would be your untreated control samples. You would compute differential expression in all your treated groups relative to those untreated animals. In your case, since it sounds like you'd like to just compare innate levels of expression between species, you need some common denominator to use as a baseline - is one of these species clearly the most common of the 9 in the habitat(s) you collected from?

2. In order to have any statistical significance of differential expression, you must have actual biological replicates. Without them, you have no way to determine the inherent variance of expression within each species, and without knowing that variance, you simply cannot compute statistical significance of your sample relative to your baseline.

So, were these species collected as pooled samples of several individual amoeba each? If so, did you run multiple sets of pooled amoeba from each species (at least 3, preferable 5 or more in order to capture the range of inherent intra-species variability)?
__________________
Michael Black, Ph.D.
ScitoVation LLC. RTP, N.C.
mbblack is offline   Reply With Quote