For the cluster.split command I’ve been using:

cluster.split(fasta=current, count=current, taxonomy=current, splitmethod=classify, taxleve=4, cutoff=0.15)

For some reason it changes the cutoff from 0.15 to 0.01. There is no error message, it just changes it. Because of this when I do the make.shared command I can’t group it to 97% similarity. All my OTUs are now 99% similarity and we have a ton of extraneous OTUs.

We ran it twice and it did it both times. We then changed the pre.cluster diffs to 3, this removed about 25% of our OTUs, but once again, when we ran cluster.split the cutoff was once again changed from 0.15 to 0.01 without error message or explanation.

I’m using the latest version.

Any suggestions?

You can try increasing the cutoff value.

Hi, I had this exact same problem and tried working without a cutoff- always crashed since one temporary file that was made was just way too big; but I did tinker with cutoffs and using 0.25 worked for my data set.