Hi, I had a silly doubt regarding alpha diversity.
I have 2 samples with me and both of them are healthy control samples of the gut microbiome. My doubt is can i get alpha diversity indices for them individually ?
Hello,
Did you already tried to run core-metrics plugin? If yes and it was successful, your alpha diversity metrics per sample should be in files named "{metric}_vector.qza".
Also, in "available plugins" section of Qiime2 docs you can find commands for calculation of individual alpha diversity metrics by your choice without running core-metrics plugin.
Hi Timur
Thanks a lot for your reply. I didn't run the core metrics plugin but i ran certain commands for certain alpha indices individually. I am getting my qza files for individual alpha diversity indices (chao, shannon and pielou to be precise) but i am not able to get the corresponding visualization files (qzv ones). I presume there are some issues with my metadata file which are beyond my understanding at present.
Hi!
Probably, your metadata is fine. Visualization of alpha-diversity metrics is available when group-significance is tested. Since you only have 2 samples, you can not perform such analysis. But you should be able to export (check exporting tutorial) alpha metrics as a tsv file to check indices for each sample.
Thanks for the response Timur...
I am done with that but if you could shed some light on "Since you only have 2 samples, you can not perform such analysis" I previously was able to generate diversity indices and visualization files for individual samples from a group of 20 samples. So does the number of samples involved plays a role in it ? Also is there a minimum cutoff in terms of number of samples to gain fruitful diversity results.
Really appreciate your inputs and prompt response. Means a lot
Thanks
It is possible to perform stat analysis (I guess you used alpha group significance plugin) with 20 samples, but 2 samples would not be enough. But it is only related to the visualization in Qiime2. To calculate alpha diversity, the number of samples should not affect indices itself. As I already mentioned, those indices can be checked by exporting alpha diversity artifacts to tsv files.
Alpha diversity is always estimated for one and single sampling unit. By its definition: it is the diversity of one and single sampling unit. There is no problem with estimating the alpha diversity for one (or two) sampling units, but there may be self-inflicted problems of further handling those alphas.
This topic was automatically closed 31 days after the last reply. New replies are no longer allowed.