Normalized counts


#1

I have read about the how normalized counts, but I don´t understand why my total final counts were differents, i.e., if MEGAN normalized with the smallest counts samples, the final counts in all my samples should be the counts of smalles counts, but this is not true.

I normalized six samples with de next final counts (without normalized)
128493 185242 157920 126678 106733 120368

smallest sample

and when normalized the counts are
102224 95403 96325 105014 106733 102451

even two samples lost aprox. 50 % of counts.

I would like to know why this happens?

Thank you soo much