# Cluster.split by classification

**URL:** <https://forum.mothur.org/t/cluster-split-by-classification/1652>\
**Category:** Commands in mothur\
**Created:** [January 21, 2014, 3:57am UTC](https://forum.mothur.org/t/cluster-split-by-classification/1652 "2014-01-21T03:57:08Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![tw123](https://avatars.discourse-cdn.com/v4/letter/t/3da27b/32.png) [@tw123](https://forum.mothur.org/u/tw123)\
**Post date:** [January 21, 2014, 3:57am UTC](https://forum.mothur.org/t/cluster-split-by-classification/1652/1 "2014-01-21T03:57:08Z")

</div>

Hi Pat and Sarah,

Apologies if this has already been asked but I am wondering how the taxlevel argument works in the cluster.split function. If I have a bunch of seqs classified with many of them at order level being unclassified but spanning different classes and phyla, if I set the taxlevel to 4 would it group those unclassified sequences as 1 ‘split’ or would it consider the higher orders of taxonomy?

Also, as an aside, is there a reason cluster and cluster.classic no longer have the option to specify large=T? is this because of the cluster.split?

Thanks,  
Tris

---

<div class="post-metadata">

**Author:** ![pschloss](https://yyz2.discourse-cdn.com/flex036/user_avatar/forum.mothur.org/pschloss/32/4_2.png) [@pschloss](https://forum.mothur.org/u/pschloss)\
**Post date:** [January 22, 2014, 12:48pm UTC](https://forum.mothur.org/t/cluster-split-by-classification/1652/2 "2014-01-22T12:48:31Z")

</div>

Hi Tris,

It would consider the higher orders - so if you have a bunch of order level “unclassifieds” but they come from different classes, then they would be split into separate groups by class.

As for large=T, I’m not sure what is going on there. However… we’ve found the large=T option to be a disaster for the average neighbor algorithm. It ends up sucking up more memory than the large=F option and so it’s hard to imagine a case where you’d want to use large=T.

Pat