Do not run multi.spiec.easi with unidentical sample scheme

[TeeratatK]

I work with phyloseq objects. I have a problem with a number of replicates that are not the same (one has 5 and another has 4) for making cross-domain analysis. How to deal with data like this?

Error in spiec.easi.list(datalist, method = method, sel.criterion = sel.criterion, : Do not run multi.spiec.easi with unidentical sample scheme

[zdk123]

Measurements have to be drawn from the same samples in order to perform any kind of association/correlation analysis. If you have an extra sample, you can drop it, but only four samples is really small for this kind of approach anyway.

[TeeratatK]

I'm sorry, my question might make you confuse. I work with 3 phyloseq objects (5 replicates for each, total of 15 samples) and use for creating cross-domain analysis. But one phyloseq object has only 4. I think it causes an error above for this reason. Can you recommend how to overcome this problem?

[zdk123]

Same answer. The number of samples and same names must match between all of the objects for cross domain measurements. There is no built- in way to subset the samples for inter- vs intra- domain objects. I could consider adding that but you'd have to do that yourself given the current implementation

On Sun, Apr 7, 2024, 12:54 PM TeeratatK @.***> wrote:

I'm sorry, my question might make you confuse. I work with 3 phyloseq objects (5 replicates for each, total of 15 samples) and use for creating cross-domain analysis. But one phyloseq object has only 4. I think it causes an error above for this reason. Can you recommend how to overcome this problem?

— Reply to this email directly, view it on GitHub https://github.com/zdk123/SpiecEasi/issues/257#issuecomment-2041528780, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAUD2RCMAIEOHSEQEX77EL3Y4F25JAVCNFSM6AAAAABF3EQG4GVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDANBRGUZDQNZYGA . You are receiving this because you commented.Message ID: @.***>