faster way to generate metadata tsv file

[moonlight (John)]

Hello, I have a bacterial 16S rRNA dataset. I have created a metadata tsv file initially, before I started to QC and analyze the data.

Obviously, I need to filter out the samples with low number of reads, rarefy the the feature tables etc. which may lose some of samples.

I would like to created a new metadata tsv file that only includes the samples that pass the QC and read to analyze. – this is new metadata tsv file is a subset of the initial one.

Is there any way that I could do this in a faster way? (not manually). Otherwise, I have check the number of samples in each step.

Thanks.

[thermokarst]

Most actions in QIIME 2 that take metadata are configured to automatically work with a superset of samples, for precisely this reason - you often wind up dropping samples as an analysis progresses.

@moonlight

Is there any way that I could do this in a faster way?

Try leaving your sample metadata alone and see if you even need to subsample your metadata file - my bet is you won't need to.