0.2.0a1 (2015-08-13)¶

changes since 0.1.7 (2014-05-10)

Bug Fixes¶

Removed previous Ensembl data; <0.3% of previous sequences were inconsistent with newer ensembl releases

Added liftover mechanism to bootstrap new schemas from existing data
Standardize on “NCBI” as origin name
When RefSeq transcript exon structures differ between chromosome and patch versions, use chromosome version [0926cf34e173]
Use explicit sequence sources when loading sequences
Bumped schema version to 1.1 for addition of associated_accessions and tx_similarity_v

New data: ensembl-79, refseq 2015-07-31
Added RefSeq NR (non-coding) transcripts
Added geneacs (associated accessions)
Added transcript similarity view [40a4e734d606] with multiple similarity metrics (experimental)
Added Change Log to documentation
Developed and documented release flow w/tools
Implement gene-based data subsets for testing loading
Explicitly set cds_{start,end}_i to None when not defined in txinfo (for non-coding transcripts) [e82011d1b02c]

Added uta subcommand and make rule to update materialized views [b1f2ef185320]
Add load-origin from tsv file; update ensembl Makefile rules [dc421df5ef19]
Added “mega-test” and “the-whole-kielbasa” test targets
Added indexes to tx_def_summary_mv [c1cda1cfd4c9]
Enforce FK relationship from exon_set.tx_ac to transcript.ac [ad5309895b5c]
Facilitate restartability in ensembl-fetch by creating canonical gene order and batching fetches
Standardize on schema name as uta_1_1 for dev, then rename on release
Updated code to use new eutils [#171, d831e02f51fa]
Update to use bioutils instead of bdi [769eea9241e4]
Updated to use the aligner from uta-tools-align (f.k.a. locus-lib-bio) [a1f492a32f5e]