How can I perform learnErrors on a subset of reads?

General Discussion
1

Should I directly downsample using the nreads or nbases parameters, or should I downsample each sample first and then use the learnErrors function?
image
If use the nreads or nbases parameters and samples are sequenced quite deeply, the program will only use a very small number of samples, so it takes a long time to run.should I downsample each sample first and then use the learnErrors function?

1 Like
2

Hello @bio-Tao,

Could you provide some more context? Which software/tool is this in reference to?

1 Like
3

I'm sorry, I just saw your message because I was on vacation recently. I used Dada2 in Qiime for denoising (method: denoise-ccs).

4

Hello @bio-Tao,

It sounds like you are using the dada2 R software directly (calling functions from the R package (learnErrors), not using the qiime2 software)? If this is the case then dada2's GitHub will be the better place for more involved support.