Amptk

- bug fix in `ufits taxonomy` if `--tax_filter` was used the filtered OTU table would not be correct in the BIOM output file
- fix for `ufits ion` if using `--mult_samples` now creates mapping_file correctly.
- updates to the docs on new usage

Several major changes in this version:
- taxonomy for ITS is updated to newest release from UNITE v7.1 11-20-2016.
- USEARCH9 is now supported throughout and defaults have been changed to use `usearch9`
- UNOISE2 algorithm is employed in a 'clustering' module
- SINTAX algorithm is supported in `ufits taxonomy`. default hybrid method now uses SINTAX, UTAX, and global alignment to infer the best taxonomy assignment.
- QIIME-like mapping files can now be used during demultiplexing/pre-processing. If you do not use a mapping file, the scripts will create one for you. The mapping file can be used to add metadata to it and then passed to `ufits taxonomy` to create a BIOM output file containing all metadata, OTUs, and taxonomy
- BIOM output of `ufits taxonomy` is compatible with QIIME, PHINCH, MetaCoMET, PhyloSeq, etc.
- `ufits filter` now alerts user if passing a barcode name via `-b` is not found in OTU table

- updated UFITS with better logging for external programs, so now log file should be more informative if you run into any errors. this will help me diagnose the problem.
- bug fix for `ufits dada2` where script would die if `--uchime_ref` option not passed

- update to COI database. Previous version had some mistakes in re-formating the BOLD database. Scripts and workflow on how this database was constructed is available [here](https://github.com/nextgenusfs/ufits/blob/master/docs/reference_databases.md)
- updates to `ufits dada2` pipeline. Script will now also create bOTUs (biological OTUs) as the DADA2 output is sensitive to 1 bp, thus a single "species" may be spread out over several iSeqs. Therefore, to accommodate downstream community ecology statistics, these iSeqs are clustered at a set threshold (`-p, --pct_otu`) to collapse "species" into OTUs.
- updates to `ufits data2` so that it builds an OTU table in same manner as `ufits cluster`, i.e. original reads are mapped to iSeqs (as opposed to DADA2 only using quality filtered data for OTU table generation).

- added support for reference chimera filtering in the `ufits dada2` OTU picking method

- improve the terminal output of `ufits dada2` as well as the Rscript logging