=====
* Added 'resource_report' and 'phylogeny_report' utilities.
* Additional reporting for phylogeny pipelines:
o 'genetree' reports maximum likelihood tree when run on a supermatrix.
o supermatrix image in 'multalign', ordered by most complete taxon and gene.
o some histograms were changed to tables for small numbers of taxa.
* Updates to README and TUTORIAL:
o Clarified that the Agalma-bundled SwissProt database only includes Metazoa.
o Fixed overwrite of 'BIOLITE_RESOURCES' variable in TUTORIAL. (24)
* 'homologize' now ignores bad BLAST hits, that seem to occur for query
sequences longer than 10Kb and in which the original query id is lost in
the output.
* Fixed bug with passing flags through to RAxML in 'genetree'. (19)
* Removed a hard-coded minimum cluster size of 3 from 'multalign' and replaced
with the 'min_taxa' value (which should never be less than 4).
* New mechanism to break up the expensive all-by-all tblastx in 'homologize',
so that many smaller chunks can be run externally/concurrently, and read
back into the pipeline. This feature is not yet tested and we plan to finish
it in the 0.3.3 release.
* Fixed default RAxML model in genetree. (9)
* New regression test feature 'agalma test' downloads and runs a small
transcriptome and phylogeny example to verify correct installation and
validate changes to the code base.
* Phylogeny pipelines can now pass a common ID with --id and they will
intelligently find the appropriate output from earlier pipelines. Previously,
numeric run IDs had to be passed between pipelines. This is demonstrated in
the TUTORIAL.