OTU clustering pipeline for next-gen reads

 
About otupipe
The otupipe script creates OTUs from next-generation sequence reads for single-region experiments such as 16S and ITS. Otupipe performs error-correction, amplicon and abundance estimation, chimera detection (using UCHIME) and OTU clustering. See here for OTU benchmark results. Otupipe requires USEARCH.

Version 1.1 is designed for reads generated by Roche 454 sequencing of a region of a gene extracted from a fixed pair of primers. Typically the gene is 16S, but this script should also work well for other regions such as ITS. It is not designed for shotgun reads or for reads generated by an Illumina sequencer. Quality filtering is a major problem with Illumina. If you have Illumina data, then I suggest keeping only the largest OTUs, i.e. those containing at least something like 5k or 10k reads. 

Otupipe is a bash script, so you must have a bash shell installed. If you're using Windows, you can run otupipe under Cygwin.

The manual was updated Sept 30th 2011 with comments on using otupipe with Illumina.

Downloads (v1.1.9)
Source code.
Manual (PDF).
Gold database.
readmap2qiime (Python script to convert readmap.uc to QIIME OTU table).