How to simultaneously analyze 16S for bacteria and ITS for fungi

General Discussion
1

Hello, I am working with biofilters inoculated with bacteria and fungi, with samples taken at different times. The sequencing was done for 16S (bacteria) and for ITS (fungi). I have carried out the analyzes in Qiime2, for the two populations separately, reaching the analysis of alpha and beta diversity. Is it possible to compare the content of fungi and bacteria within the samples? Can I combine these analyzes to see the proportions of bacteria and fungi within the samples? I have been searching the forum but have not found any help in this regard. I would appreciate any guidance you can give me.

2

Hi @conscc ,
Good questions!

I am not 100% sure what you mean here, but I think the answer is no, not within QIIME 2. You could look at correlation networks of bacteria/fungi, but this would need to be done outside of QIIME 2 as there is not a suitable plugin for this at the moment.

On the other hand, there are other comparisons that you could make within QIIME 2. Some examples:

  • Procrustes analysis and mantel tests (both available in the q2-diversity plugin) would allow you to compare beta diversity ordinations/distances based on bacterial and fungal data, to see how closely they correspond.
  • alpha diversity measurements could probably be correlated directly
  • individual species of interest could also be compared directly if you want to focus on a specific hypothesis. E.g., q2-sample-classifier could be used to see if bacterial composition is predictive of the relative abundance of a specific fungal ASV or species or alpha diversity (or vice versa).

No. The data were sequenced separately with different marker genes, and the read counts do not correspond to absolute abundances, so by combining bacterial and fungal data you could not make any meaningful conclusions, e.g., that one is more abundant than the other.

good luck!

1 Like
7

Hi Nicholas, thank you very much for your answer. Now it is clear to me that I cannot do a joint analysis by sample. I'll explore your suggestions, which will take some time, since I'm new to the area. I hope to have good news later. Greetings.
Cony

1 Like
8

Hi,

Sorry, I'm late to the party, and I hope it's okay I step in.

I don't have a QIIME 2 solution, but you might look at the work of Laura Tipton. She's done some interesting network analysis in this area that you might find useful.

Best,
Justine

1 Like