Four dodecamers: tiny peak (peak 2) height

I know this is going to sound like I dont care about traditional statistical methods, but in this case I makes NO sense to me at all to include many 0 in a mean as a missing value when the results puts the peak height below the background.  SO, while i have calculated that missing value in on the old database, and the new database (and the only one i will continue to get peak heights from) and shown them both here.  I have taken just the values from instances where this tiny peak (peak 2 – adjacent to the N term peak) occurs.  Those means are presented in the bottom screen print from excel. I also will stop reporting about the normal distribution (see medium green cells), which can happen when data of missing values are excluded) otherwise that makes no sense to me either.