core_diversity_analyses without sampling depth

[m. salomez]

Hello Qiimer,

One question regarding the workflow core_diversity_analyses. I would like to generate an OTU rarefaction plot having all my samples with all the sampling depth for each of the sample.

I have tried to remove the sampling depth -e in the command but it is not possible. Someone know how to do to have a complete rarefaction plot?

regards

Melanie

[Jose Antonio Navas Molina]

Hi Melanie,

The script that you want to run is alpha_rarefaction.py. This is the script that core_diversity_analyses uses under the hood to generate the rarefaction plots, but you'll have more freedom choosing the parameters.

Hope this helps!

[m. salomez]

Hello Jose,

Thank your for your help. I will try with this workflow.

Melanie

[m. salomez]

Hello again,

In this workflow, as the default value we have :

-e, --max_rare_depth

The upper limit of rarefaction depths [default: median sequence/sample count]

I would like to do the analysis without rarefaction depths step. I still don't know how to do that. I would like to have a rarefaction plot with all my samples and all my reads.

Melanie

[Jose Antonio Navas Molina]

Hi Melani,

I may not be understanding what are you trying to accomplish. I think you are trying to generate alpha rarefaction plots like this one  but it doesn't end at a given rarefaction depth, it stops once you've used all your reads, correct?

In such case, change the default value of "-e, --mak_rare_depth" to be the maximum number of sequences that you have in a single sample. You can check this number by summarizing the BIOM table.

Cheers,

[m. salomez]

Hi Jose,

I would like to have a rarefaction plot like this one. Instead, I have a rarefaction plot with the cut-off of the sampling depth like this one.

The script discards the samples with less than the value (e) of the sampling depth. For example, If I set the sampling depth with 3000 Otus, all my sample with less than 3000 Otus will not appear on my graphic. I would like the whole samples to appear on the graphic.

Regards

Melanie

[Jose Antonio Navas Molina]

Hi Melanie,

If you run the command alpha_rarefaction.py providing the parameter -e with the value being the maximum number of sequences in a single sample in your table, it will generate the plot that you're looking for (I just tested on some test data).

Cheers,

[m. salomez]

Hi Jose,

Thank you very much. It is working with this command.

Melanie Salomez

[Jose Antonio Navas Molina]

Glad it worked!