Normalization method for alpha diversity

[iordanis (H20)]

Dear all,

I have some questions about normalization methods.

I have a relative abundance table from qiime2 outputs, i want to calculate the alpha diversity via vegan package.

so here is my question, relative abundance are the TSS normalization method?

If yes can i use this to calculate alpha diversity?

[colinbrislawn (Colin J Brislawn)]

Hello @iordanis,

Normalizing is surprisingly contentious!

It has been discussed at length on the forums:

Use of rarefacted Table

Hi @m_max, Welcome! There's been a fair bit of discussion around rarefaction here (or maybe Im just involved in a lot of them). It's maybe good to go look into those threads since they'll give you a wide variety of view points. Here are a few to start that I found pretty quickly. Best, Justine

What published papers have you read that you found most convincing?

[iordanis (H20)]

Hi @colinbrislawn,

i am sorry for the late response and thank you for the help.

I read out papers that suggest the relative abundance is a normalized method, like TSS normalization. Although for meta-analysis the authors suggests that TSS normalization is better than rarefaction.
Those papers are lie maaslin2 MMUPHin and reviews for statistical methods!

I have one more question for different abundance test with ANCOM with qiime2.

For ANCOM analysis we use the absolute abundance or relative abundance table?

Best,
Jordan

[colinbrislawn (Colin J Brislawn)]

Maaslin2 and MMUPHin are from the Huttenhower Lab. I've worked with them before!

Note that both methods are proprietary to Harvard:

biobakery/MMUPHin/blob/master/LICENSE

biobakery/Maaslin2/blob/master/LICENSE

ANCOM needs raw data. It does its own normalization internally!

[iordanis (H20)]

So, do you know which method use relative abundance table, for differential abundance analysis?

Yes maaslin2 and MMUPHin are powerful tools for my opinion!!

[colinbrislawn (Colin J Brislawn)]

@iordanis

So, do you know which method use relative abundance table, for differential abundance analysis?

I worry there has been a slight miscommunication here. If I wanted to test for differential abundance, I'm not sure I would recommend relative abundance anymore...

Have you read this paper?

Microbiome Datasets Are Compositional: And This Is Not Optional

[iordanis (H20)]

To be honest, no i did not read this paper?

And this is a little bit confound for me because when i use the MMUPHin tool the tutorial suggest as input at the genus lvl of taxa the relative abundance table.

However thank you for this paper i will read it!!!!

[iordanis (H20)]

Dear @colinbrislawn,

I reiterate my gratitude for yours support and the paper that you told me to read.

So i will provide you another review which explain why we can use the relative abundance table or another methods for abundance table.

Either way, the method used depends on the biological question. If I have misunderstood something, please correct me.

https://www.nature.com/articles/s41522-020-00160-w

Best,
Jordan