Documentation on post processing of simulated sequences

[ahfitzpa]

Could you add additional documentation on the recommendations for post processing of the simulated sequencing data.

Following simulation of sequencing data, fastq or fastq.gz files were filtered and trimmed using X,Y and Z.

For example, typical ONT data is basecalled and demultiplexed using Guppy or Bonito. If the user includes adapters and barcodes in the sequences, should they demultiplex with guppy to permit filtering of sequencing WITHOUT a barcode on either end. Or should the user rely on more generic methods such as NanoFilt and cutadapt/trimmomatic?

[rrwick]

I think it all depends on what you're planning on using the reads for.

As an example, if you ultimately just want adapter-free reads, it doesn't make sense to have Badread add adapters and then trim them off afterward - you could just tell Badread to not add adapters. But if you wanted to test a pipeline that included adapter trimming, then having Badread add adapters might be good.

Regarding barcodes, Badread can't create barcoded reads in a single run. If you wanted to test a demultiplexing pipeline, you would have to run Badread multiple times, each with a different barcode, and then combine the results.

Does that all make sense? Let me know if I can elaborate!