TODO

1. Compare the accuracy of UMI Reducer to similar tools.
2. Generalize UMI Reducer to work with existing pipelines and multiple types of sequencing data. 
3. Modify "extract reads" wiki to be more friendly to computer scientists.
4. Modify collapsePCRduplicates.py to accept Genome Research format reads.
5. Write discussion of how to customize the system requirements of UMI Reducer