technical question using envfit with ordinate phyloseq function! #666
Unanswered
cabraham03
asked this question in
Q&A
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Hi. I have a technical question.
I want to generate a NMDS plot along with the OTUs based on the envfit statistics. If I have less than 3,000 OTUs I don’t have any problem, but if I have a >19,000 OTUs my machine just simply stop or get slow with envfit.
So, I want to generate a NMDS ordination using ordinate phyloseq function, and then obtain the distribution of the Samples ($points) and the distribution of the Species ($species), and then make a envfit analyses and used the significant (< 0.05) to filter and plot them together, but using the distribution generated with ordinate function.
I will use GlobalPatterns as examples, its a good example because it present >19,000 OTUs
extract Samples (points)
now Species
To avoid problems, I will generate a filter and then use envfit to obtain the significant
now envfit
obtain the significant
then filter the significant OTUs from the Species generated with ordinate function
and then generate a plot !!
My question is, if I have so many OTUs is valid to filter and then obtain the statistics with envfit, and generate the distribution with ordinate phyloseq function ? I mean, the direction, and length of the arrow in Species is representative of the relatedness with the Samples distribution using this method ?
I know that I can generate the the plot with for Samples and Species obtained from the filter data (GP1) and generate the ordinate and envfit, but I rather to use all OTUs to generate the Sample Distribution !!!
any suggestion ?
Thanks
Beta Was this translation helpful? Give feedback.
All reactions