But after I assembled the trimmed reads, I tried to upload the assembly to the TSA database in NCBI, they gave me the error saying that my sequence is contaminated by primer sequences. I found one of the contamination sources using vector screen, which is ‘CCCTACACGACGCTCTTCCGATCT‘. But this sequence is actually contained in one of the adapter sequences in the list:
Trimming does not look for all subsequences of the adapter. It only detects the adapters from their start and then continuing towards the end (at variable lengths). Normally this is the way adapters show up. In your case it seems more oddities are present.
Trimming accuracy varies in different trimmers. I’d recommend atria to determine and trim the adapter sequences. It is a newly-published cutting-edge trimmer with exceptional precision and speed.
To find a concise trimming benchmarks, you can click here.
You can also find more comprehensive trimming benchmark at Atria’s paper.