-
Notifications
You must be signed in to change notification settings - Fork 8
Home
GuilhemSempere edited this page Jan 25, 2024
·
21 revisions
Latest version may be tested online at https://gigwa.southgreen.fr/
Source code repository for v2.x: https://github.com/SouthGreenPlatform/Gigwa2
v2.7 is available at [Gigwa homepage](https://github.com/SouthGreenPlatform/Gigwa2/releases/tag/2.7-RELEASE) as of january 25th, 2024:
Changes in v2.7:
- Support for up to 10 groups of individuals
- New export formats : vcf.gz (bgzipped), BayPass (allele counts by population, i.e. group)
- Instance contents summary table available from a link on the homepage
- Ability to add a custom informative section to the homepage (see customHomepageParagraph property in online docs)
- Ability to find gene names via a lookup widget
- Flapjack-Bytes UI improvements
- Added support for individuals with multiple samples in the same run
- If a single sequence range is selected in Gigwa filters, directly focus on it when opening IGV.js
- Simplified the genotype format in GFF3 export
- Various bugfixes
v2.6.1 is available on Gigwa homepage as of may 30th, 2023:
Changes in v2.6.1:
- Added means to view/update project description from administration interface
- Changed chart interval size default and max values to respectively 200 and 500
- Better support for filtering non-SNP variants by position (now accounts for end-sites)
- Fixed a query bug (introduced with v2.6
⚠️ ) leading some filter combinations to erroneously return 0 result => we advise to activate the below setting to clear possible wrong cached counts - Added "clearCachedCountsOnStartup" config property
v2.6 is available on Gigwa homepage as of may 5th, 2023:
Changes in v2.6:
- Improved filtering: MAF, missing data and heterozygosity percentages may now be applied by range
- New export functionalities for converting SNP genotypes into pseudo-alignments ready for tree construction (supported formats: NEXUS, PHYLIP, FASTA)
- Added means for any user to push exported data into Galaxy libraries (for any Galaxy instance the user has an account on)
- Import interface refactored to allow providing metadata along with genotypes (single-shot). Now attempts pulling BrAPI metadata as soon as extRefId and extRefSrc values are found (no more need to get back to the import interface and submit again)
- Added support for sample metadata (only accessible via BrAPI)
- Implemented the following BrAPI v2.1 calls: GET /allelematrix, POST /search/allelematrix, PUT /calls
- Importing genotypes via BrAPI v1 now prompts for token rather than doing unsecure authentication
- Improved support for online output tools, allowing to finely define URLs (in config.properties) for pushing each type of exported file according to its extension
- Added support for allowedOrigins parameter in config.properties, providing sysadmin means to precisely define CORS-policy based on URL patterns (convenient for securing REST API usage)
- Allowed to override config.properties values with environment variables by prefixing their name with "GIGWA." (particularly useful when using the Docker version)
- Various bugfixes
v2.5.1 is available on Gigwa homepage as of august 26th, 2022:
Changes in v2.5.1:
- At genotype import time, support for sample / individual mapping file, providing ability to consider provided biological material names as samples rather than individuals
- Support for sample metadata (may be uploaded or pulled via BrAPI)
- BrAPI V2: added consistency to germplasm-related calls, implemented additional sample-related functionalities
- Flapjack-Bytes now able to display unmapped variants
- Various bugfixes
v2.5 is available on Gigwa homepage as of april 1st, 2022:
Changes in v2.5:
- Additional import formats: Intertek, Flapjack
- Ability to select markers by ID lists
- Embedded IGV.js, along with tens of remotely available genomes
- Enhanced version of Flapjack-Bytes (faster loading, more features implemented)
- Dynamic statistical analysis graphs: Fst, Tajima’s D
- Single-Sign-On compatibility via C.A.S. implementation
- Database dump / restore facility
- Enhanced BrAPI v1 & v2 metadata pulling functionality (working with germplasm & sample entities)
- Significant speed improvements for imports & exports
v2.4.1 is available on Gigwa homepage as of september 3rd, 2021:
Changes in v2.4.1:
- Importing using as many threads as available cores (faster!)
- Now storing sequence length when provided in VCF (and this able to re-export it)
- Fixed a major bug (introduced in v2.4) making variant genotypes appear several times when exporting from multi-project databases
- Solved miscellaneous minor issues
v2.4 is available on Gigwa homepage as of august 6th, 2021:
⚠️ Don't use v2.4, it contains a bug: please use v2.4.1 instead!
Changes in v2.4:
- Speed improvements in variant search, density chart calculation and (especially) export procedures
- Ability to export individuals' metadata (when available) along with genotyping data
- Made group discrimination feature available regardless of selected genotype patterns
- Integrated Flapjack-Bytes prototype for online visualization of datasets exported into Flapjack format
- Allowed feeding individual's metadata from multiple BrAPI (V1) sources at a time
- Implementation of BrAPI V2 calls covering pretty much the entire Gigwa data model
- Migrated Swagger interface to Springfox3
- Showing host name for each existing database in the administration section
- Various bugfixes
- Upgraded several prime Java dependencies (Spring-data, MongoDB driver, htsjdk...) and third-party software for bundles (JRE v8u291, Tomcat v8.5.66, MongoDB v4.2.15)
NB: As of v2.4, in Docker images and bundles, default WiredTiger compression algorithm is zstd rather than zlib (configured in startGigwa.* scripts)
v2.3 is available on Gigwa homepage as of march 25th, 2020:
New in v2.3:
- Added feature to import individual metadata from BrAPI v1 sources (via germplasm + attributes related calls, see documentation for details)
- Added an experimental BrAPI v2 server implementation
- Implemented additional BrAPI v1 calls: GET germplasm, GET URL_GERMPLASM_ATTRIBUTES
- Added Docker containers (see docker-compose.yml)
- Added an updateGigwa script to bundles, that comes in handy to upgrade the web application without affecting configuration files
- Added unit tests to make sure not to break query engine when implementing new features
- Various bugfixes
- NB1: As of v2.3 the --directoryperdb option is set by default when launching MongoDB. Please refer to https://docs.mongodb.com/manual/reference/configuration-options/#storage.directoryPerDB if you need to migrate existing data (alternatively, you may disable that option in docker-compose.yml for Docker or in the startGigwa script for bundles)
- NB2: As of v2.3, in bundles, tomcat and mongodb binaries' folders are named just "tomcat" and "mongodb" (no more version numbers). This will help keep startGigwa / stopGigwa scripts consistent across future releases.
v2.2 is available for download on Gigwa homepage as of july 15th, 2019:
v2.2 provides much faster hapmap import and allows any authentified user to add his own metadata! By upgrading MongoDB version from 4.0.4 to 4.0.10, it also becomes compatible with curl4
v2.1 is available for download on Gigwa homepage as of may 17th, 2019:
v2.1 fixes various bugs and features a query-bookmarking functionality! It also allows to define import and export size limits more conveniently (refer to maxImportSize and maxSearchableBillionGenotypes properties in the docs: for both of them a default setting may now be used in conjunction with settings specific to each user)
v2.0 is available for download on Gigwa homepage as of february 27th, 2019:
Because of modifications in the data structure, v2 is incompatible with data imported using previous versions
v2 is faster, more robust, and adds lots of new functionalities. Documentation is now embedded in the web-application
Gigwa v1 wiki is available at https://github.com/SouthGreenPlatform/gigwa/wiki