Yes, you are very right. But the reason I want to do this is not that I want to evade the bioinformatic analysis, but try to find a cheaper and ideal way to fulfill my goal. Since I am dealing with allo-multiple-ploidy species, it is tedious to clone individual homeologs or homologs, and then sanger sequence them, but it would be too costly to use whole transcriptome/genome sequencing, since I am just interested in that specific gene. So what was on my mind is similar to what you suggested, I will barcode the sequence I want with unique sequence for each sample, and I pool all my samples into one library, so I can use just one deep sequencing to do all my samples. I want to know if there is anyone who has done anything similar before and has a mature protocol that I can follow.