# can I run anosim for interactions (two factors)?

**URL:** https://forum.qiime2.org/t/can-i-run-anosim-for-interactions-two-factors/27096
**Category:** General Discussion
**Tags:** permanova
**Created:** [July 17, 2023, 4:50pm UTC](https://forum.qiime2.org/t/can-i-run-anosim-for-interactions-two-factors/27096 "2023-07-17T16:50:01Z")
**Posts on this page:** 6
**Page:** 1

<div class="post-metadata">

### Author: ![MarwaTawfik](https://forum.qiime2.org/letter_avatar_proxy/v4/letter/m/ee59a6/32.png) [@MarwaTawfik](https://forum.qiime2.org/u/MarwaTawfik)
#### Post date: [July 17, 2023, 4:50pm UTC](https://forum.qiime2.org/t/can-i-run-anosim-for-interactions-two-factors/27096/1 "2023-07-17T16:50:01Z")

</div>

Continuing the discussion from [Adonis vs. Anosim vs. Permanova](https://forum.qiime2.org/t/adonis-vs-anosim-vs-permanova/9744/5):

I checked your advice regarding the interaction of two factors using permanova, I used ANOSIM in my analysis so could I just stick to it in the interaction analysis also?  
Thanks in advance

 ![image](https://forum-qiime2-org.s3.dualstack.us-west-2.amazonaws.com/original/3X/9/b/9b9e0a9385a4570679d5442340de891b1d98521c.png)  
ks

---

<div class="post-metadata">

### Author: ![cherman2](https://forum.qiime2.org/user_avatar/forum.qiime2.org/cherman2/32/17129_2.png) [@cherman2](https://forum.qiime2.org/u/cherman2)
#### Post date: [July 26, 2023, 4:38pm UTC](https://forum.qiime2.org/t/can-i-run-anosim-for-interactions-two-factors/27096/4 "2023-07-26T16:38:24Z")

</div>

Hi @MarwaTawfik,  
Could you elaborate on your question?

As you quoted, if you have multiple factors Adonis is probably the best approach for testing.

---

<div class="post-metadata">

### Author: ![MarwaTawfik](https://forum.qiime2.org/letter_avatar_proxy/v4/letter/m/ee59a6/32.png) [@MarwaTawfik](https://forum.qiime2.org/u/MarwaTawfik)
#### Post date: [July 31, 2023, 11:09pm UTC](https://forum.qiime2.org/t/can-i-run-anosim-for-interactions-two-factors/27096/5 "2023-07-31T23:09:46Z")

</div>

Many thanks for your response @cherman2 Chole  
do you know when to use anosim generally speaking?  
I used adonis (PERMANOVA) for general effects and interaction but I preferred to use ANSOIM for pairwise comparisons as it is more sensitive to location variation and correlation. What do you think?

Thanks in advance for your input.  
Cheers  
Marwa

---

<div class="post-metadata">

### Author: ![cherman2](https://forum.qiime2.org/user_avatar/forum.qiime2.org/cherman2/32/17129_2.png) [@cherman2](https://forum.qiime2.org/u/cherman2)
#### Post date: [August 3, 2023, 12:06am UTC](https://forum.qiime2.org/t/can-i-run-anosim-for-interactions-two-factors/27096/6 "2023-08-03T00:06:33Z")

</div>

Hi @MarwaTawfik,  
Here is a discussion on research gate that might be able to answer your question: [https://www.researchgate.net/post/Should-I-use-ANOSIM-or-PERMANOVA-to-analyses-community-differences-within-an-estuary](https://www.researchgate.net/post/Should-I-use-ANOSIM-or-PERMANOVA-to-analyses-community-differences-within-an-estuary)

Hope that helps!  
🐢

---

<div class="post-metadata">

### Author: ![MarwaTawfik](https://forum.qiime2.org/letter_avatar_proxy/v4/letter/m/ee59a6/32.png) [@MarwaTawfik](https://forum.qiime2.org/u/MarwaTawfik)
#### Post date: [August 3, 2023, 1:27pm UTC](https://forum.qiime2.org/t/can-i-run-anosim-for-interactions-two-factors/27096/7 "2023-08-03T13:27:17Z")

</div>

That helps, thanks so much, Chole. @cherman2

---

<div class="post-metadata">

### Author: ![MarwaTawfik](https://forum.qiime2.org/letter_avatar_proxy/v4/letter/m/ee59a6/32.png) [@MarwaTawfik](https://forum.qiime2.org/u/MarwaTawfik)
#### Post date: [August 3, 2023, 1:28pm UTC](https://forum.qiime2.org/t/can-i-run-anosim-for-interactions-two-factors/27096/8 "2023-08-03T13:28:16Z")

</div>

I think I could use both PERMANOVA and ANOSIM, the former for general effects and interactions and the latter for pairwise comparisons