All notable changes to this project will be documented in this file.
The format is based on Keep a Changelog.
- Re-release for dependency updates
- Certain errors in the search service now return error messages instead of a generic 500.
- The
cigar
field returned by the/families/{id}/relationships
endpoint is now in the CIGAR format, not just CIGAR-inspired.
- Fixed sorting by "type" or "subtype" fields in
/families
- FASTA sequences submitted for sequence search with blank sequence names (i.e. ">\nACGT") are assumed to have the name "Input", "Input_1", etc., instead of failing to return any results.
- "Annotations" search results that overlapped the requested range but did not completely fit inside it were not included if they were on the '-' strand.
/families/{id}/annotation_stats
: New endpoint, providing the data from two endpoints in one query:/families/{id}/assemblies
and/families/{id}/assemblies/{assembly_id}/annotation_stats
.
- Fixed calculation of species on
/taxa/coverage
endpoint.
/families
(only in the "full" format) and/families/{id}
include three new fields:source_method
,source_method_description
, andsource_assembly
. Seeapi/swagger/swagger.yaml
for more details./families/{id}/seed
now reports the average kimura divergence of the alignment if available./families
and/families/{id}/relationships
have a new filter:include_raw
/families/{id}/relationships
has a new filter:include
/taxa/coverage
: New endpoint, reporting a count of leaf species which have Dfam entries assigned to them.
- Some fields will be filled in with an accession where a "name" is usually specified, if the "name" is not available. This is done when the field is intended to be human-readable.
- Results for the
/families/{id}/relationship
endpoint are now limited to 300.
- "Method not allowed" errors respond with a 405 error instead of 500
- "Annotations" search now includes hits that overlap the requested range, in addition to hits that are fully contained within the range.
- Fixed a regression that caused
/families
withformat=full
to fail.
- Added support for single sequence downloads in 'fasta' format:
/family/{id}/sequence?format=fasta
- Added support for bulk downloads in 'fasta', 'embl', and 'hmm' formats
via
/families
. This supports the same filtering and sorting options as the existing JSON output formats ('full' and 'summary').
- Queries to
/families
will now fail if too many results would be returned from one query.
- Attempts to download invalid formats respond with a 400 error instead of a 500
- Citations will always be returned in the correct order
- New endpoint '/taxa/{id}' to retrieve information on a single taxon
- Family objects now include a 'version' field
- Updated handling of seed region names and positions
- Nonexistent search jobs are properly caught and handled as a 404
- Fix stockholm formatter seed regions with duplicated positions (e.g. chr:1-5:1-5)
- Fix /families classification filter matching more than it was supposed to
- Use the search start date instead of job start date to find search results
- Use the SQL charset that matches our databases
- Initial public release of the Dfam API.