rarefaction curve with number of samples on x axis

Hello everyone,

I have a question regarding rarefaction curves.

When using the 'qiime feature-table rarefy' command, we obtain a graph that displays Sequencing Depth on the x-axis.

My question is as follows: is it possible to generate rarefaction curves in qiime2 that show the number of samples on the x-axis?

I would like to create this plot to demonstrate that my sample size has reached sufficient coverage.

Thank you for your help

Hi @Edoardo_Scali,
Just for clarification, I believe that the method you are referring to is qiime diversity alpha-rarefaction and not qiime feature-table rarefy. qiime diversity alpha-rarefaction produces a visualization like you are describing and qiime feature-table rarefy produces a rarefied table with no visualization.

We dont have a plot that allows for visualizing number of samples on the x-axis, but the second plot in that visualization shows number of samples on the y-axis and sequencing depth on the y axis so that you can see if you are losing too many samples at specific sampling depths. Does that sound like what you are looking for?

If not, Can you elaborate on what you are trying to investigate? I will try to help you find a visualization that will work!

Hope this help!

1 Like

Hi @cherman2,
First of all, thank you very much for your kind response.

What I'm trying to do is to produce rarefaction curves with the number of samples on the x-axis and the number of ASVs on the y-axis. Additionally, I would like to be able to visualize multiple curves on the same graph, for example, one curve for each unique value contained in one of the variables I have in my sample metadata.

I tried using "qiime diversity alpha-rarefaction", and I had a chance to confirm your description and what I see in the .qzv file.

If it's not possible to obtain the graph I would like to produce in qiime2, are there alternative methods for generating and visualizing rarefaction curves that allow for high-quality graphics that can be included in scientific publications?

Thank you very much for your help! :grinning:

Hi @Edoardo_Scali,

It sounds like you are looking for a fully featured graphing library!

There are quite a few to choose from, so here are some that are popular:
https://seaborn.pydata.org/ (Python)
https://ggplot2.tidyverse.org/ (R/tidyverse)
Plotly · GitHub (JS / Python / R)

As with all things, it's a trade-off:

Qiime2 <---> graphing library
simple complex
fast slow(er)
consistent flexible

Thanks @colinbrislawn for the help!


This topic was automatically closed 31 days after the last reply. New replies are no longer allowed.