Skip to content

Release 2025.2.0#455

Merged
bencap merged 178 commits intomainfrom
release-2025.2.0
Jun 13, 2025
Merged

Release 2025.2.0#455
bencap merged 178 commits intomainfrom
release-2025.2.0

Conversation

@bencap
Copy link
Collaborator

@bencap bencap commented Jun 11, 2025

This release adds support for ClinGen Allele ID (CAID) linkages in MaveDB, the storing of Base Editor data, the ability to link ClinVar controls to mapped variants (via CAID) and the ability to support multi-target and accession based score sets via our mapping process.

Features

Closes #311, Closes #314 and Closes #381: #388
Closes #317: #395
VariantEffect/dcd_mapping2#2 and VariantEffect/dcd_mapping2#3: #436

Enhancements

Closes #350: #420

Bug Fixes

Fixes #432: #433
Fixes #441: #442
Fixes #61: #447

bencap and others added 30 commits March 29, 2025 10:57
Refactors dataframe validation logic into 3 component files: column.py, dataframe.py, and variant.py.
This simplifies the validation structure and logically separates validation function based on the part of the df they operate on.
Refactors most of the test suite to better identify dependency separation problems. Validation tests
may now be run with only core (and dev) dependencies installed, and fixtures which operate on server dependencies are conditionally loaded based on the installed modules.
With this change, it will be much more straightforward to identify dependency 'leaks', or server dependencies which mistakenly are leaked into validation type code.
This allows the use of the vs-code pytest extension but still prevents the use of external connections. Enabling this socket makes it easier to test within the code editor.
The hgvs package is not able to parse allelic variation (multi-variants denoted by brackets), which
are often a key variant string in base editor data. We work around this by:
- Parsing the multi-variant into MaveHGVS without any target info to ascertain whether it is syntactically valid
- Parsing each subvariant against the provided transcript to ascertain whether it is informationally valid
Adds tests for multi-variant validation for accession based variants. As part of this change,
an additional transcript was added to tests genomic based protein variants in addition to
just testing nucleotide based variants.
Prior to this, we weren't really using SeqRepo to do transcript resolution (unintentionally). Note that to use SeqRepo in this manner, a new environment variable `HGVS_SEQREPO_DIR` should be set.
bencap and others added 24 commits May 28, 2025 12:05
…ontributor

Modified add_contributor function and add two more related tests.
…s-path-to-score-range-data-model

Add OddsPath Property to ScoreRange Data Model
variant_mapper_manager previously failed because of an incorrectly placed
'where' function and because we did two related db queries, one for a score set
and one for target gene(s) within that score set, which resulted in an error
when adding the score set changes to the db. Instead, query to select the score set,
and then loop through the score set's target genes rather than querying the target genes
table in the db.
…SetsAndExperiments

Bugfix/estelle/61/num score sets and experiments
Support mapper update for multi-target score sets
@bencap bencap added the core: release A release PR label Jun 11, 2025
@bencap bencap requested a review from jstone-dev June 11, 2025 19:06
@bencap bencap merged commit 0c1b391 into main Jun 13, 2025
8 checks passed
@bencap bencap deleted the release-2025.2.0 branch June 13, 2025 01:01
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

core: release A release PR

Projects

None yet

4 participants