Skip to content

Vidjil-server

Note

This changelog concerns the Vidjil web application, client and server. As we are using continuous integration and deployment, some features are pushed on our servers between these releases.

Web 2024-01⚓︎

Milestone 2024-01

Complete list of issue and change of this release can be found here.

Main change of this release is based under the hood, with conversion of backend server from an old framework web2py to a new implementation py4web. Mains ideas are shared between these frameworks, yet it required a lot of hard work.

We also changed the docker-compose file to adapt it to new docker services. Be aware when you make migration to this release to have a good understanding of change in this file. See docker changelog to get more information.

Other improvments of client and server are limited.

Improve analysis * New axis with size of clonotype inside his locus. Also add an associated preset Size within each locus (17).

API, https://www.vidjil.org/doc/api * API have be adapted to new version of server. If you want to use it with previous web2py version, please use script in web2py-stable branch. * Now a group number can be specified. If your account is member of only one group, the selection will be done automatically. If you are member of multiple groups, a list of them will be print and you will have to select which one you want to use.

Bugs corrected * AssignSubset link if now fixed

Continuous integration * A complete rewriting of unit tests on server have been done.

Web 2023-03⚓︎

Milestone 2023-04

Complete list of issue and change of this release can be found here.

Improve analysis

  • New view displaying warnings and allowing to select related clonotypes
  • New preset "locus by sizes"
  • New .csv export of metadata information of a sample
  • Custom primer sets can be set on local server installations

Improve reports and report templates

  • Shared template across analyses
  • Harmonize rendering of clonotypes depending of the number of samples
  • Better selection of loci
  • Better clone name display, using N region parameters

Improve ergonomy

  • Remember columns and feature layers selection in aligner
  • Keyboard "Esc" shortcut to close open panels

API, https://www.vidjil.org/doc/api

  • More endpoints: impersonating, analysis download, data deletion with a confirmation

Bugs corrected

  • Reflect changes on grid on reports
  • Resolve error on the downloading of result file
  • Fix "number of samples" axis
  • Resolve bug in selection of sample by presence of selected clonotypes
  • Cleaner error message on broken files
  • Faster groups controller
  • Automatic deletion of temporary (and potentially large) files from addon scripts

Continuous integration

  • More testing on Docker images (tools, germlines)

Web 2022-06⚓︎

Milestone 2022-06

Complete list of issue and change of this release can be found here.

Improve analysis

  • More sample information: diversity measures by germline, summary of warnings
  • More clonotype information: detailed productivity
  • More tags and colors, new color by sample
  • V gene identity computed by IMGT now takes into account ins/del events
  • Distributions clonotypes can now be filtered by locus

Improve ergonomy

  • New reports, with flexible template/blocks
  • More flexible inclusion of third-party sequence analysis tools
  • Many fixes, including better page resets between analyses

Improve patient/run/set database usage

  • Search can be done by set number
  • Upload from network includes a search field
  • Returning to a sample page keeps the current analysis configuration

Interoperability, API

Improve software quality

  • New cypress tests on client and tutorial (deleting old watir)
  • Updated documentation, also with a search field

Other points

Web 2021-11⚓︎

Milestone 2021-11

Complete list of issue and change of this release can be found here.

Improve analysis

  • Primers can be shown on the sequences and be removed before sending sequences to external tools such as IMGT or IgBlast
  • AIRR external analysis can be uploaded ("AIRR/.vidjil/.clntab import")

Improve ergonomy

  • Add button to launch analysis on all files of a patient/run/set
  • Renamed some analysis axes
  • Some axes have now sliders

Improve internal database usage

  • Faster access to databases (patient/set/run lists)

Improve traceability

  • New logs on pre-process steps, such as for merging ("i" on samples)
  • Access logs are available again (on the public server, since mid-2019)

Improve software quality

  • New tests and improved test pipelines (cypress), improved Docker containers

Other points

Web 2021-05⚓︎

Improve analysis

  • New sequence aligner, flexible display of nucleotide/AA sequences, VDJ, CDR, and FR features
  • New detection of primer positions and computation of interpolated length (BIOMED-2 and EuroClonality-NGS primer sets)
  • New report and axis, detailed cause of non-productivity

Improve ergonomy

  • Better handling of saved settings (color by, naming choices)
  • New button to open one sample with its analysis from the patient/sample database
  • Color settings are used in report

Improve quality testing

  • The tutorial is better tested, including interactions with the sample database

Other points

  • Fix disgracious position in graph
  • Add more explicit warning about healthcare compliance
  • Fix logout error for some instances (app.vidjil.org)

Web 2020-12⚓︎

  • Possibility to create at once multiple patients/runs/sets by pasting data from the clipboard

Add new warnings

  • Add warnings when the server is not approved for healthcare data (!813, see >vidjil.org/doc/healthcare>)
  • Add warning to show variation in V(D)J designation or productivity prediction between samples (#4566, #4578)

Improve quality testing

  • The tutorial is now tested, step by step (#2880)

Improve internal database usage

  • Refactor some queries of the database. Filtering and searches are now way faster. Note that the basic query may be a little bit slower. (#3169)

Other points

  • Fix error on clusterBy (#4495)
  • Many other bug fixes

Web 2020-07⚓︎

Better study full repertoires

  • Genescan-like distributions on the full repertoire (preset 4) (!595, !551, !525)
  • New configuration to export the full list of clones in AIRR format (with no clone limitation)

Clone list

  • New options to lock clone sorting (!630, !564)

Time graph, working with several samples

  • Better options to hide/show samples in the time graph (!565)
  • Possibility to hide clones not present in the current sample (!564)
  • New design of curves

Other points

  • Better organization of configurations (!627)
  • Free text zone in report
  • Read merger: Flash instead of PEAR (#3911)
  • Many bug fixes

Web 2020-01⚓︎

  • New behaviour of graphList (#4105); Now you can hide/show sample directly from a list with checkbox, and button for automation

You can also hidding them by double clicking on their label in the timeline graphic.

Web 2017-10⚓︎

  • New tag mechanism to store/filter default or custom metadata on patient/runs/sets #2170 #2172 #2683
  • Added a second grid view when there is only one sample #2244
  • New experimental feature to add germline genes in the segmenter #1925
  • New axes with number of deleted nucleotides #2506
  • Better test process (jshint, nightmare.js) #2272 #2273
  • Moved continuous integration to Gitlab CI (.gitlab-ci.yml)
  • New and updated unit and functional tests
  • Bugs corrected: Fasta export #2628, values on axes #2540, spurious data #2598, timeout in segmenter #2559...

Web 2017.05⚓︎

  • Refactor of the axes for the grid and bar plots. New features are forthcoming #1471 #1763 #2175
  • New option to download output files #2141 #2405
  • Refactor of the sequence view. The sequence view will soon be able to display V(D)J germline genes.
  • New setting to display shorter clone names (by default, allele is hidden when it is *01) #1679
  • Renamed "merge" to "cluster", more consistent clustering options, including undo #2336 #2335
  • New keyboard shortcuts on some browsers #1227
  • Made Docker configuration easier to update #2469
  • Improved CSV export with additional information on productivity and CDR3 #2428
  • Updated results on sample deletion #1631
  • Bugs corrected: #2304, #2371, #2386, #2402, #2466, ...

Web 2017.03⚓︎

  • New server installation option to get data from mounted directories instead of uploading them #1588
  • New "color by clone" option #2205
  • New "hide clone" option #2227
  • Faster server processing, multi-threaded #1919
  • New paginated output of the sampleset pages #2037
  • Removed limit of 16 displayed samples into a set #2269
  • Bugs corrected: reports #1979, #2281, IMGT segmenter #2273