Amptk

- some bug fixes and updated docs
- fixed the local blast search for `ufits taxonomy`
- fixed `ufits filter` to work with any mock community fasta file
- update `--uchime_ref` to be able to use custom database
- fix logging in `ufits database` command

A somewhat major bug fix: the de-multiplexing script for Ion Torrent data was not removing reverse primer correctly. This has been fixed.
- complete re-write of the `ufits filter` script to now normalize counts per number of reads in each sample prior to running the `--index_bleed` filter to remove noise from the dataset.
- `ufits filter` script now can deal with synthetic mock spike-in control (set as default). Calculates index-bleed in both directions and smartly filters OTU table. (email me for details on synthetic mock if you are interested).
- Taxonomy database was updated to most recent UNITE release
- Taxonomy database installation method is changed, now downloads pre-formatted databases for ITS1, ITS2, and FULL length ITS sets.
- Updated UCHIME reference sequences
- New functionality for `ufits taxonomy` that allows for removal of non-fungal OTUs with the `--only_fungi` argument
- Added script for sub-sampling or rarefaction of data prior to clustering
- Added script for selecting or removing samples from a de-mulitplexed data set
- Added script to append an OTU table to meta data file
- Added support for FUNGuilds functional annotation of taxonomy in OTU table.

- improvement of `ufits database` command for processing the taxonomy information for training UTAX
- added support for illumina data that is in a single file, i.e. with similar read setup as 454/ion `ufits illumina2`. I've seen this type of data from MrDNA service
- updated `ufits install` command to by default train UTAX for full length, ITS1, and ITS2 for `ufits taxonomy`
- minor bug improvements

- update to support Roche 454 reads (SFF or FASTA/QUAL as input) and need a fasta file of barcodes used
- minor bug fixes
- `ufits install` now skips primer trimming for UNITE+INSD database as it takes a lot of time without any added benefit because global search uses full length of the OTU and db doesn't matter if it is trimmed or not

- Several changes to syntax for commands to make easier to use.
- Added an `install` option to download, format, and create taxonomy databases from UNITE
- by default `ufits taxonomy` now uses a hybrid approach (UTAX and USEARCH) to get most out of taxonomy from legitimate hits. If USEARCH hit is < 97% identical, then UTAX is used. If hit is > 97% then taxonomy from UTAX vs USEARCH is compared and whichever result has more levels of taxonomy is used.

- minor bug fixes
- update to include `ufits summarize` command that generates taxonomy level OTU tables as well as makes a stacked Bar graph of each level of taxonomy in the dataset.