It sounds like you are trying to create a set of non-redundant sequences. There are many methods to do this but the important thing to consider is how many sequences you have because not all methods are equal in this regard. Uclust may work just fine, though if you have a lot of sequences (e.g., one or more lanes of HiSeq) I don't think this will work out well, in my experience. I recommend you try Vmatch if you have access to it (there is a section in the manual on how to accomplish this task).
The proper name of what you are asking for is "repeat masking". So, you have to look for tools to "mask" low entropy sequences.
There are many tools to mask sequences: