NMDS Ordination, 3D Plots, PC-ORD File

[Christine Prasse]

Hi there!

I am using QIIME for the first time to analyze my 454 data.  Thus far I have successfully produce the PCoA plots using the QIIME tutorial.  My new objective is to create NMDS ordination plots.  I was originally going to convert the OTU file to a matrix compatible with PC-ORD, but I read in the forum I could run the NMDS ordination in QIIME.  I ran the nmds.py script using the otu_table.txt files produced from jackknifed_beta_diversity.py script:

nmds.py -i jackknifed_beta_diversity_6000/weighted_unifrac_otu_table.txt -o jackknifed_6000_unweighted.txt

nmds.py -i jackknifed_beta_diversity_6000/weighted_unifrac_otu_table.txt -o jackknifed_6000_weighted.txt

I selected the weighted_unifrac_otu_table.txt and weighted_unifrac_otu_table.txt to run the nmds.py script because they were equivalent to the  output files produced by beta_diversity.py script.  I am interested in making plots so I passed a single file through the make_3d_plots.py (as advised to a previous user) using the following command:

make_3d_plots.py -i jackknifed_6000_weighted.txt -m 454_Map_SY2

 

The script ran, but it did not produce a plot (see attached).   I read thorough the other 3d options, but I don't necessarily know where to start because the script ran instead of producing an error.  

My questions for you:

1. Is it possible to make the 3d NMDS ordination plots using the make_3d_plots.py command?  Will this script produce similar plots as PCoA?  
2. I would also like to run some other analyses in PC-ORD; however, the user that mentioned "that it was fairly easy to convert OTU tables to PC-ORD compatible files" did not include instructions on how to perform this task.  Could you please provide the script or direction so I could transfer the file?

Thank you for your time.

Christine

[Will Van Treuren]

Hi Christine, 

Can you zip the output directory for this command 'make_3d_plots.py -i jackknifed_6000_weighted.txt -m 454_Map_SY2' and send it?

Thanks,

Will 

Christine

--
 
---
You received this message because you are subscribed to the Google Groups "Qiime Forum" group.
To unsubscribe from this group and stop receiving emails from it, send an email to qiime-forum...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.
 
 

[Christine]

Hi Will,

Here's the zip output file you requested. 

Christine

[Will Van Treuren]

Hi Christine, 

There should be an output directory, not just a single html file. Can you send the entire output directory? I can open the file you sent in the King viewer, but the axes percentages are all 0 and I am generally unclear on what I am looking at. 

> 2. I would also like to run some other analyses in PC-ORD; however, the user that mentioned "that it was fairly easy to convert OTU tables to PC-ORD compatible files" did not include instructions on how to perform this task.  Could you please provide the script or direction so I could transfer the file?

What format are PC-ORD files in? QIIME does not contain a convert to PC-ORD functionality that I know of, but if its a simple otu counts X samples matrix then convert_biom.py is what your looking for if you are trying to move from .biom to otu counts X samples matrix. 

Will 

[Christine]

Hi Will,

I'm sorry for the confusion. Here's the zipped folder created when I run the 3d_plots script.  As for part 2, let me research the original posting and I will get back to you.

Christine

[Will Van Treuren]

Hi Christine, 

I can only open the king file, but one of my colleagues is able to open your .html file with chrome, so I am not sure whats going on. I would suggest viewing the king plot (which you will find in the randomly named folder in the 3d_plots folder you sent) with the King viewer which can be download here. After discussion with this same colleague it looks like the NMDS plot is correct (the axes shouldn't have any variance explained). 

Best, 

Will  

[Christine]

Hi Will,

I did not use the King viewer to review my data.  I have been able to resolve the second question on my own.

Thank you for your help.

Christine