Bigsdb

Minor formatting changes.

In this version:

* Added assembly checks - results of these can be queried and used for filtering of isolate datasets.
* Added announcement banner to database front page. The contents of /etc/bigsdb/announcement.html will be shown on all database front
pages - this is useful for service announcements, e.g. scheduled maintenance.
* Upgraded D3 to v6, and changed charting library from c3.js to billboard.js.
* Added means of grouping standard isolate fields into categories. Separate groups are listed in dropdown boxes in queries and plugins, and on the isolate display page.

This version adds new functionality for assessing and filtering results based on genome assembly metrics.

N50 and L50 values are now stored and can be used in isolate queries. The stats are automatically updated by a database trigger whenever, and however, contigs are added to the sequence bin.

%GC, number of Ns, and gaps in an assembly can be calculated and stored by an external script after assembly uploads. These can be used along with N50, total length and number of contigs in isolate queries.

The results of arbitrary analyses can now be stored in the isolate database as JSON values. These can be displayed within an isolate record using a templating system. rMLST species checks have been implemented as an exemplar.

Checks are now performed on submissions prior to upload to the database. Basic assembly stats are available to submitters and curators with automated warnings shown based on customisable thresholds. Hard limits can also be set so that submissions can be rejected automatically.

We can also run the rMLST species check on submitted genomes so that the results are available to curators shortly after submission and prior to handling.

Batch uploading data is now performed in a separate process, with progress notifications loaded by AJAX polling. This prevents browser timeout when hundreds of genomes are uploaded together.

This version adds an option to limit the number of running jobs for particular plugins.

This version introduces annotation status based on scheme completion.

For a well-annotated genome it can be expected that all the loci, or a specified threshold number of loci, for some schemes will have allele designations. We can use this annotation status metric to help assess genome quality. Individual schemes can be set to be used for quality metrics and threshold numbers of loci that indicate a 'good' or 'bad' annotation based on that scheme can be defined in the schemes table.

The metric can be used in searches to filter datasets that have passed annotation criteria.

This release provides minor bug fixes.