The Laplace value allows to select the best number of communities obtained by the Dirichlet approach. However, sometimes the Laplace values of several K are very close and it would be interesting to see the results of these other partitioning.
Is there a way to obtain the results of these “unselected” partitioning?
In advance, thank you very much for your help!
Bernard
The results for the grouping in each iteration can be found in the *.posterior files. It takes a bit of work, but you can translate those into design files for each grouping scheme using any spreadsheet program.
Thank you! I’ll have a look.
Great support, as usual.
Bernard
In the *.posterior file, I guess that the closes value to 1 tells you to which partition the sample belongs. Right?
Yes, for each OTU/row the 1 tells you which partition it’s in.