Alpha and Beta Diversity analysis using shotgun metagenomics data

[Synapse (MJ)]

Dear all,

Good day QIIME community!

I am new to QIIME2 and would appreciate some guidance. We are working on a project involving a biofilm, and we plan to send our samples to a facility for shotgun metagenomic sequencing. I have read about the Moving Pictures and Gut-to-Soil Axis tutorials and how they can be used for diversity analysis with QIIME2 Amplicon. However, we cannot use these tutorials because they are only used for amplicon data. I have some questions about our circumstances and humbly request for your guidance.

1. Using the QIIME2 MOSHPIT toolkit, is it possible to perform alpha diversity analysis (Shannon, observed features, Faith's PD, evenness, etc.) and beta diversity analysis (Jaccard, Bray-Curtis, unweighted UniFrac, and weighted UniFrac) using shotgun metagenomics data? If so, is there a guide, and what would the workflow look like? I read the MOSHPIT tutorials, but I was unable to find information about alpha and beta diversity analysis (although I might be wrong and just misunderstood the contents).
2. If diversity analysis using QIIME2 MOSHPIT is not possible, what alternatives can we use? Is there something we can do with the shotgun metagenomics data to perform analysis using QIIME2 Amplicon?

Thank you very much!

[timanix (Timur Yergaliyev)]

Hello and welcome to the forum!

@Synapse

Using the QIIME2 MOSHPIT toolkit, is it possible to perform alpha diversity analysis (Shannon, observed features, Faith's PD, evenness, etc.) and beta diversity analysis (Jaccard, Bray-Curtis, unweighted UniFrac, and weighted UniFrac) using shotgun metagenomics data? If so, is there a guide, and what would the workflow look like? I read the MOSHPIT tutorials, but I was unable to find information about alpha and beta diversity analysis (although I might be wrong and just misunderstood the contents).

Yes, you can calculate alpha and beta diversity metrics, excluding ones that require phylogenetic tree (faith pd and unifracs).

1. Use qiime2 moshpit for taxonomy annotation with kraken2.
2. Switch to amplicon distribution and use resulted feature table for non-phylogenetic core-metrics, which will provide you with Shannon, eveness, number of features, Bray-Curtis and Jaccard metrics.

@Synapse

If diversity analysis using QIIME2 MOSHPIT is not possible, what alternatives can we use? Is there something we can do with the shotgun metagenomics data to perform analysis using QIIME2 Amplicon?

You can use any pipeline for taxonomy annotation, then collapse counts to desired taxonomy level, import to qiime2 and calculate non-phylogenetic metrics.

For one project, I used metaphlan4 for taxonomy annotation and then modified provided by them phylogenetic tree to calculate phylogenetic metrics, but it required some scripting to modify the files.

Hope that helps.

Best,

[Synapse (MJ)]

Hello, @timanix.

@timanix

Yes, you can calculate alpha and beta diversity metrics, excluding ones that require phylogenetic tree (faith pd and unifracs).

1. Use qiime2 moshpit for taxonomy annotation with kraken2.
2. Switch to amplicon distribution and use resulted feature table for non-phylogenetic core-metrics, which will provide you with Shannon, eveness, number of features, Bray-Curtis and Jaccard metrics.

@timanix

You can use any pipeline for taxonomy annotation, then collapse counts to desired taxonomy level, import to qiime2 and calculate non-phylogenetic metrics.

For one project, I used metaphlan4 for taxonomy annotation and then modified provided by them phylogenetic tree to calculate phylogenetic metrics, but it required some scripting to modify the files.

Thank you for this! I think I understand the gist of it. I will try it with some sample shotgun metagenomics data to better understand it. Thank you very much!