should barcode in scRNAseq be deleted when demultiplexing

Question

[Binghong]

I demultiplex fastq files using the demultiplex program ,and find the result didn't delete barcode, I want to ask if it is necessary to delete the barcode in RNAseq. any suggestion is appreciated!

[Binghong]

the barcode is in the first 8bp of sequence in read 2

Answer 1

[benformatics]

If you have already demultiplexed your samples I see no reason for you to keep the barcode information. Unless you plan to do some kind of barcode error correction (e.g. hamming distance collapsing). Having a random 8-mer present in your sequences will definitely bias any downstream alignments. Again, assuming that this read 2 is being aligned to the genome/transcriptome.

[Binghong]

thanks for your apply, I want to ask if the different length of read 1 and read2 will influence the analysis that follows when the barcode is deleted from read2

[benformatics]

The effect should be minimal.

[Binghong]

thanks for your apply!