Metadata preparation and rarefaction level

Hello @MMC_northS @thermokarst and all ....thank you so much ,,, Now I am able to do import and able to follow the tutorial.

But I have a few doubts again -

  1. While doing alpha-diversity, it required metadata.tsv file. Kindly can you explain how should I prepare this file, if you any example can you share with me?

  2. Sampling depth how should I decide, option says we should use total frequency but which frequency should I use. When I checked my table.qzv file, I found "total frequency" "mean/min/max frequency per sample" and "mean/min/max frequency per feature" ?

Here is the image for metadata and sampling depth "

". Kindly help me to solve these issues.


Hi @khemlalnirmalkar

  1. Regarding the metadata-file, this is the file called in QIIME1 “mapping-file” and is a tsv file that must have the columns #SampleID BarcodeSequence LinkerPrimerSequence and Description, and this last should be at the end of the file. In the middle you can write all the columns that you want to analyze with your data: pH, temperature, SampleType, etc. and they can be categorical or numeric values.

Your have information about it and one example in this two parts of the QIIME2 information. Probably you should read them.

  1. Regarding to the sampling-depth depends on your data. If you use one number lower than the minimum reads that you have in one sample, that sample will be not consider in the analyses. In the table.qzv file you have the information that you told me and other visualizer tab called “Interactiva sample detail” where you can know the number of reads per Barcode, Description, etc. If you have in Description one identifier name of each sample you can know how many reads you have per sample to calculate.
    in addition, make alpha rarefaction to know how variable are you samples depending on the depth is useful too.

Good luck!


Thanks so much for the info, @MMC_northS!

I just want to point out here that those column requirements are from QIIME 1, and are not required columns in QIIME 2 (QIIME 2 has no column requirements at all, besides the index column, which should be your sample IDs or feature IDs). Thanks, and keep on QIIMEing! :t_rex:


This topic was automatically closed 31 days after the last reply. New replies are no longer allowed.