How many replications are required for QIIME2 analysis(ANCOM as q2-classifier and q2 longitudinal etc)?

[Aqleem12 (Virtue)]

Dear All,

I have used the following sample metadata.tsv for analyses of eight samples. There are two subjects. Each subject is different from other. Each sample was different from other. However there were no replications. You may have seen the moving picture tutorial the gut was repeating more than three times i.e. gut. gut. gut. The body site category of each subject has four stages. Each stage is unique however no replication. You can see here;

The genesis performed well without replications. When I started to use differential abundance testing with ANCOM however got the following message at the final command:

I donot have any replication for performing analyses further. Now I cannot send the samples to the company again. I know statistically the data will not be valid. However If anyone could suggest me what kind of strategy should I adopt to analyses this type of data with Qiime2. Now I donot have any more samples to send to the company. Secondly what parameters or commands should I need so that I may complete my analyses and may get somehow results. I am so confused. If someone help me I would be thankfull. The charges of sequencing are also costly.

Thanks

With best Regards,

Aqleem Abbas

[Nicholas_Bokulich]

Hi @Aqleem12,

You must have replicates to perform any statistical test. This is an issue with your experimental design, not with QIIME 2 itself; unfortunately, there is nothing that we can do to help you with this issue.

You can still report your results as qualitative observations (e.g., show barplots of taxonomic composition in each sample), but you will not be able to make any conclusions about whether microbiota changed over time or in response to any treatments.

Adequate replication is necessary for any statistical method. In the future, I would recommend getting assistance from a statistician to plan your experimental design and ensure that you have adequate replication for the statistical tests that you plan to use. The level of replication and types of replication really depend on the tests that you plan to perform and the biological questions that you hope to answer, so are really well out of the scope of this forum.