We are looking to run the make.biom command using data that we have normalized using DeSEq2. The way the file was done we had to transpose the .shared file to read into R and then normalized it. Once done, we transposed the data back into the original orientation and added the label column back in. The output from R is in .txt or tab delimited because that is what the .shared file is formatted with (or so it appears). However, when I place the file back into mothur and use the make.biom file as seen below, it is not reading the file correctly to make the .biom. We are getting an output from mothur that it calls .biom but it is only recognizing one of our sites rather than all 24.
I did a simple trial of resaving the .shared file as a .txt and then trying to run that same file using the make.biom and get the same issue. So it is my guess that the file needs to be structured in a certain way. Therefore my question is, is there any way to recreate the .shared file from a .txt file?
Any help troubleshooting what may going arise or any ideas would be helpful.
If you would like to use R to normalize, mothur is expecting the shared file to be in the following format:
//this first line the “label Group numOtus” are expected. The rest of the line must have the same number of labels as numOtus.
label Group numOtus Otu01…
//608 is the big number here. Mothur will expect to read 608 number on the line after the “0.03 F003D000 608”.
0.03 F003D000 608 409 …
Most often when there are errors in shared files, it is due to a column missing. You can set the debug flag using the set.dir command to recieve additional output.
mothur > set.dir(debug=t)
If you are still having issues after that, you can send the file to mothur.bugs@gmail.com and I can take a look.