how to use --bypass-prescreen

[wuguo...@gmail.com]

Hi, 

Under what situation shell we use "--bypass-prescreen", can you give an example? Using prescreen and using a custom ChocoPhlAn database make sense, and will be consistent with the taxonomy composition results. When shell we use whole ChocoPhlAn database?

Thank you!

[Eric Franzosa]

Greetings,

You would search against the whole ChocoPhlAn database if you wanted to do a comprehensive, nucleotide-level search against microbial pangenomes. Note that under most circumstances this is "overkill" as you'd be searching against pangenomes that are not in your sample (which could also result in spurious hits). The goal of the prescreen is to isolate the subset of ChocoPhlAn that we believe is represented in your sample, which allows us to map against that subset exclusively in the downstream step.

You might choose to map against all of ChocoPhlAn if you believed that a potential species-of-interest was present in your sample but not detectable by MetaPhlAn (e.g. because it was very low-abundance), but I don't recommend starting with this approach.

Thanks,

Eric

[wuguo...@gmail.com]

Thank you, Eric !

在 2015年7月31日星期五 UTC+8上午5:01:54，Eric Franzosa写道：