Filter OTUs?

After I run the SOP on a particular batch of samples, the OTU table has 200k+ OTUs (contamination?? causes?)! Currently, I filter them out towards the end:

cluster.split(fasta=current, count=current, taxonomy=current, splitmethod=classify, taxlevel=4, cutoff=0.03)
classify.otu(list=current, count=current, taxonomy=current, label=0.03)
filter.shared(shared=current,minpercentsamples=5,minpercent=0.01)
make.shared(list=current, count=current, label=0.03)

Is there a way I can remove these much earlier in the pipeline? It would save a lot on the computation time!

thanks!

you can’t filter otus before you’ve clustered the sequences into otus. You can filter.shared before you classify.otus but not before you’ve created them with cluster.split or cluster