Extract reads where barcode was found in discovery mode

[koehlek99]

Hi, thanks a lot for developing this tool!

We're currently running flexiplex in discovery mode to identify reads with the expected read structure. We'd like to use only reads where a barcode candidate was found when running it in whitelist mode. The majority of our reads actually doesn't show the correct configuration and our whitelist is incredibly large (700 Mio.). Is there a way to write out a filtered fastq file when using the discovery mode?

Thanks a lot!

Best
Krissi

[nadiadavidson]

Hi Krissi,
The flexiplex-filter python script that is packaged with flexiplex might do exactly what you need. So your process would be:

1. Run flexiplex to get the possible barcode in your data (this doesn't give a FASTQ, just a list of possible barcodes)
2. Run flexiplex-filter to remove barcode that are not in your white-list + also empty drops/low supported barcodes
3. Run flexiplex again with the filtered barcode list. This will perform as inexact match to the barcodes and give you the final FASTQ.
   There is some documentation of this process here,
   https://davidsongroup.github.io/flexiplex/#assigning-long-reads-to-10x-barcodes-when-barcodes-are-unknown
   But please let us know if you run into any trouble or if I have misunderstood what you need to do. Also, feel free to add more detail about your particular experiment.
   Cheers,
   Nadia.