Dockerfiles, pydantic case parsing and gene variants fixes

[4.46]

Added

• Created a Dockefile to be used to serve the dockerized app in production
• Modified the code to collect database params specified as env vars
• Created a GitHub action that pushes the Dockerfile-server image to Docker Hub (scout-server-stage) every time a PR is opened
• Created a GitHub action that pushes the Dockerfile-server image to Docker Hub (scout-server) every time a new release is created
• Reassign MatchMaker Exchange submission to another user when a Scout user is deleted
• Expose public API JSON gene panels endpoint, primarily to enable automated rerun checking for updates
• Add utils for dictionary type
• Filter institute cases using multiple HPO terms
• Vulture GitHub action to identify and remove unused variables and imports

Changed

• Updated the python config file documentation in admin guide
• Case configuration parsing now uses Pydantic for improved typechecking and config handling
• Removed test matrices to speed up automatic testing of PRs
• Switch from Coveralls to Codecov to handle CI test coverage
• Speed-up CI tests by caching installation of libs and splitting tests into randomized groups using pytest-test-groups
• Improved LDAP login documentation
• Use lib flask-ldapconn instead of flask_ldap3_login> to handle ldap authentication
• Updated Managed variant documentation in user guide
• Fix and simplify creating and editing of gene panels
• Simplified gene variants search code
• Increased the height of the genes track in the IGV viewer

Fixed

• Validate uploaded managed variant file lines, warning the user.
• Exporting validated variants with missing "genes" database key
• No results returned when searching for gene variants using a phenotype term
• Variants filtering by gene symbols file
• Make gene HGNC symbols field mandatory in gene variants page and run search only on form submit
• Make sure collaborator gene variants are still visible, even if HPO filter is used

Fix gene panel edit and clarify custom inheritance models

[4.45]

Added

Changed

• Start Scout also when loqusdbapi is not reachable
• Clearer definition of manual standard and custom inheritance models in gene panels

Fixed

• Gene panel crashing on edit action

Improved gene symbol alias handling in variants filtering

[4.44]

Added

Changed

• Display Gene track beneath each sample track when displaying splice junctions in igv browser
• Check outdated gene symbols and update with aliases for both RD and cancer variantS

Fixed

• Added query input check and fixed the Genes API endpoint to return a json formatted error when request is malformed
• Typo in ACMG BP6 tooltip

New gene index in OMIM diagnoses and avoid dropping HPO and OMIM terms collection when updating database

[4.43.1]

Added

• Added database index for OMIM disease term genes

Changed

Fixed

• Do not drop HPO terms collection when updating HPO terms via the command line
• Do not drop disease (OMIM) terms collection when updating diseases via the command line

Fix missing indexes after genes update. Custom collections index building/updating

[4.43]

Added

• Specify which collection(s) update/build indexes for

Changed

Fixed

• Do not drop genes and transcripts collections when updating genes via the command line

Freeze PyMongo version, speedup creating/updating of gene panels and Phenomizer timeout problems

[4.42.1]

Added

Changed

Fixed

• Freeze PyMongo lib to version<4.0 to keep supporting previous MongoDB versions
• Speed up gene panels creation and update by collecting only light gene info from database
• Avoid case page crash on Phenomizer queries timeout

Better cancer SV filtering, stricter config file checks on case loading and other improvements

[4.42]

Added

• Choose custom pinned variants to submit to MatchMaker Exchange
• Submit structural variant as genes to the MatchMaker Exchange
• Added function for maintainers and admins to remove gene panels
• Admins can restore deleted gene panels
• A development docker-compose file illustrating the scout/chanjo-report integration
• Show AD on variants view for cancer SV (tumor and normal)
• Cancer SV variants filter AD, AF (tumor and normal)
• Hiding the variants score column also from cancer SVs, as for the SNVs

Changed

• Enforce same case _id and display_name when updating a case
• Enforce same individual ids, display names and affected status when updating a case
• Improved documentation for connecting to loqusdb instances (including loqusdbapi)
• Display and download HPO gene panels' gene symbols in italics
• A faster-built and lighter Docker image
• Reduce complexity of panels endpoint moving some code to the panels controllers
• Update requirements to use flask-ldap3-login>=0.9.17 instead of freezing WTForm

Fixed

• Use of deprecated TextField after the upgrade of WTF to v3.0
• Freeze to WTForms to version < 3
• Remove the extra files (bed files and madeline.svg) introduced by mistake
• Cli command loading demo data in docker-compose when case custom images exist and is None
• Increased MongoDB connection serverSelectionTimeoutMS parameter to 30K (default value according to MongoDB documentation)
• Better differentiate old obs counts 0 vs N/A
• Broken cancer variants page when default gene panel was deleted
• Typo in tx_overview function in variant controllers file
• Fixed loqusdbapi SV search URL
• SV variants filtering using Decipher criterion
• Removing old gene panels that don't contain the maintainer key.

v4.41.1 hotfix release

This is a hot fix patch release, correcting a crash issue with case general report generation when reports include assessments for a variant with different assessments on other cases for the same institute (same variant_id, institute_id different case_id).

Query optimizations and a few fixes

[4.41]

Added

• Extended the instructions for running the Scout Docker image (web app and cli).

Fixed

• General case report sorting comments for variants with None genetic models
• Do not crash but redirect to variants page with error when a variant is not found for a case
• UCSC links coordinates for SV variants with start chromosome different than end chromosome
• Human readable variants name in case page for variants having start chromosome different from end chromosome
• Avoid always loading all transcripts when checking gene symbol: introduce gene captions
• Slow queries for evaluated variants on e.g. case page - use events instead

Changed

• Rearrange variant page again, moving severity predictions down.
• More reactive layout width steps on variant page

Fixed case report crashing due to missing key/values

[4.40.1]

Fixed

• Variants dismissed with inconsistent inheritance pattern can again be shown in general case report
• General report page for variants with genes=None
• General report crashing when variants have no panels
• Added other missing keys to case and variant dictionaries passed to general report