Can you explain what you are confused about?
This command will allow you to only use 2 dimensions, make sure to use the --p-number-of-dimensions parameters!
Thank you so much, Chloe!
I tried --p-number-of-dimensions.
this is my 2d bray curtis, what do you think?
bray-curtis-2d.qzv (860.9 KB)
For that concern in the bloxplot, I don't understand very well why in the "distance to C" seems that there is no significant difference; the median (especially in the left graphic) seems to be not so different between C and T.
Maybe I'm wrong in interpreting the graphic. 
I'm so sorry!
Looks like that command ran successfully! 
That is definitely true that the median in the left graphic is closer then in the right graphic. I think we need to break this visualization down more to answer your question.
What are you investigating when C is compared to itself?
What would you expect the value of a group compared to itself to be?
What would it mean if that value was relatively high, like we see in the left graphic?
I hope these questions can help point you in the right direction.
3 Likes
Thank you very much Chloe and sorry, perhaps my questions are inappropriate, but this is the first time I have approached these statistical analyses.
I expect the control has not different compared to itself, while it is different from the treated. What I don't understand is that looking at the graph on the left for example, there is "distance from control" and below on the x axis we have c (control) and t (treated). So what does it mean that there are no differences in the control compared to the control and the treated but instead, looking at the graph on the right, that the treated has differences compared to the control?
What I wrote probably doesn't make sense but I don't understand.
1 Like
Hi @Linda_Abenaim
I think you are on the right track! Lets keep digging into this visualization 
Yes, this is correct. Typically. we would expect that C compared to C would be more similar then C compared to T.
So When we are looking at the the left box plots you are looking at the distance to C. The C box plot shows the Control compared to itself and it generally looks relatively high, It seems like the control group is pretty different from eachother,
On the right plot, we are looking at distance to T. The T box plot shows that T compared to T is more similar then C compared to T. T seems to be relatively similar when compared to itself.
I believe this is what is giving your statistical power, that T compared to T is pretty similar to itself and C compared to T is not very similar. But it additionally give you some knowlegde that C seems to be a pretty varied group, while T seems to me more similar to itself.
Does that make sense?
I hope this helps! Let me know if you have any more questions!
2 Likes
thank you so much Chloe!! this was so helpful 
1 Like