hello, Recently i got paired end data and firstly, i removed the adopter sequences. For that i used fastx_clipper but i got the number of reads are different in R1 and R2. So i searched the solution for that on biostars and i found that ppl first merging the R1 and R2 files then do filtering for removal of low quality reads and then use splitter program for split the files.
So any one can tell me that is it correct method and one more thing i want to ask that after merging the R1 and R2 files, can i remove the adopter from merged file using Fastx_clipper then i can split the files in R1 and R2.
Please tell me which steps will be good.
Thanks