QIIME2 Vs MetaPhlAn3

Question

[Abhishek]

Hi all,

I have been going through documentation for Qiime and Metaphlan and am trying to understand the difference between the two. I do understand that Qiime is a full fledged tool that includes all steps from demultiplexing, denoising to statistical analysis and plots. Metaphlan seems to be for identifying the composition (or abundance) of individual microbes in each sample. However I do understand that taxonomy identification is possible in Qiime using a classifier from the q2-feature-classifier plugin, which means it does everything that Metaphlan does ?

To make things confusing for me, there is also a qiime metaphlan plugin (q2-metaphlan2). Is this just another option for identifying taxonomy from qiime instead of the q2-feature-classifier plugin ?

Also on checking metaphlan docs, it allows to use fastq files as input directly, which makes it good enough as a standalone tool.

I do feel that comparing qiime and metaphlan is incorrect and is like apples to oranges, but just can't get a clear idea.

A related question in qiime forum ( https://forum.qiime2.org/t/metaphlan2-vs-vsearch-for-fungal-metagenomic-shotgun-sequences/6536/4 ) does shed some light, but I'm still not quite clear.

Can someone please explain to me which of qiime2 or metaphlan I should be using to perform comparison between different groups or for obtaining relative abundance from 16s data or metagenomic shotgun sequence data ?

Answer 1

[shenwei356]

The biggest difference is qiime is used on amplicon sequencing data, while MetaPhlAn is for shotgun metagenomics data.

[Abhishek]

Thanks for the response. But in this case, how would the qiime metaphlan plugin be meaningful?

[shenwei356]

The doc provides limited hints. I guess, according to the Testing installation section, it just outputs .biom format which can be analyzed by downstream qiime2 tools.

[Abhishek]

Does that means the choice of tool is just based on data output format and not necessarily based on amplicon or shotgun ?