You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hi, I see that if you set --platform '10x-pacbio' then the default setting is --barcode_cell XC --barcode_umi XM. The flag XC only represents uncorrected barcodes, and the flag becomes CB after running isoseq correct in the pacbio pipeline (see the docs for the meaning of XC vs CB flag). I'm guessing that most users will input their data after barcode correction through the default pacbio pipeline, so you may want to consider switching the default flag to CB.
To clarify, is the cell barcode and UMI flag the only thing that changes after setting --platform '10x-pacbio', or are there other ways this affects analysis?
The text was updated successfully, but these errors were encountered:
Hi, sorry for late reply and thanks for pointing out! I have changed the default cell barcode to CB for 10x-pacbio platform.
The difference between 10x-pacbio and 10x-ont are just some formatting things like read names. No other differences that affect analysis due to parameter settings.
Hi, I see that if you set --platform '10x-pacbio' then the default setting is --barcode_cell XC --barcode_umi XM. The flag XC only represents uncorrected barcodes, and the flag becomes CB after running isoseq correct in the pacbio pipeline (see the docs for the meaning of XC vs CB flag). I'm guessing that most users will input their data after barcode correction through the default pacbio pipeline, so you may want to consider switching the default flag to CB.
To clarify, is the cell barcode and UMI flag the only thing that changes after setting --platform '10x-pacbio', or are there other ways this affects analysis?
The text was updated successfully, but these errors were encountered: