Skip to main content

v3.3.0 Release

· 3 min read

Hello all,

This release adds a significant update to the multi-variant renderer that allows it to be used for visualizing structural variants Previously, all variants >10bp were filtered out, with the idea that overlapping variants are too difficult to interpret

This PR now allows drawing large structural variants in the multi-sample variant display type. It will actually draw long, even overlapping, structural variants!

image

Figure: Screenshot showing many SV calls over a 5Mbp+ region from the 3,202 human samples from the 1000 genomes project, with a large "inversion" variant showing the shaded triangles. The right side panel shows the feature details for this inversion variant showing the new genotype frequency panel, indicating about 47% of samples have a 'heterozygous' inversion call here

image

Loading a random track, and turning on the arc display, there are indications of a large variant for a random sample here (green and navy arcs indicate discordant read pairing associated with inversion)

We hope that you enjoy these new features! The visualizations may be visually complex but we hope it can be a magnifying glass to reveal important patterns of your datasets! Feel free to let us know if you have any feedback

Downloads​

To install JBrowse 2 for the web, you can download the link above, or you can use the JBrowse CLI to automatically download the latest version. See the JBrowse web quick start for more details.

yarn run v1.22.22 $ lerna-changelog --silent --silent --next-version 3.3.0

3.3.0 (2025-04-18)​

🚀 Enhancement​

🐛 Bug Fix​

  • Other
    • #4929 Fix broken bigbed features when there is gene level aggregation (@cmdcolin)
  • core

📝 Documentation​

  • core
    • #4949 Controlling navigation of embedded component - storybook examples (@cmdcolin)
    • #4924 Add config docs for pre-processors with simplified config snapshots (@cmdcolin)
  • Other

🏠 Internal​

  • core

Committers: 1​

v3.2.0 Release

· 2 min read

Hello all,

This release includes a new ability to provide simplified snapshots to the linear genome view, a new built-in clustering method for genotypes, and more

In-app clustering​

In recent versions, we provided the ability to cluster genotype and multi-wiggle tracks, however we provided an R script for users to perform clustering externally and import the results

With this release, we have added a basic in-app clustering workflow, greatly improving the ease-of-use of this feature

image

Screenshot of the in-app clustering user interface

Renamed embedded packages and restored vanillajs builds​

We have renamed our embedded packages

  • Renamed @jbrowse/react-linear-genome-view -> @jbrowse/react-linear-genome-view2
  • Renamed @jbrowse/react-circular-genome-view -> @jbrowse/react-circular-genome-view2
  • Renamed @jbrowse/react-app -> @jbrowse/react-app2

Motivation: our vanillajs builds, which people often pointed to the "latest" release on from CDN networks, were no longer able to be updated in a backwards compatible way.

Therefore, we have renamed the packages, and updated tutorials.

This allows us to continue updating the package without any breakage for existing users. Please let us know if you run into any issues!

Downloads​

To install JBrowse 2 for the web, you can download the link above, or you can use the JBrowse CLI to automatically download the latest version. See the JBrowse web quick start for more details.

yarn run v1.22.22 $ lerna-changelog --silent --silent --next-version 3.2.0

3.2.0 (2025-03-22)​

🚀 Enhancement​

  • core
    • #4919 Add in-app clustering method for multi-wiggle tracks (@cmdcolin)
    • #4909 Add ability to load refNameAliases from chromAliases.bb file (@cmdcolin)
    • #4906 Add in-app clustering method for genotype matrix (@cmdcolin)
  • Other
    • #4914 Allow looking up header column by name for RefNameAliases (@cmdcolin)
    • #4910 Allow refNameColumn in RefNameAliasAdapter to override the default displayed refnames (@cmdcolin)
    • #4907 Simplified snapshot API for initial navigation of LGV (@cmdcolin)
  • app-core, core, product-core, web-core
    • #4899 Add aborting to multi-variant rendering, multi-wiggle rendering (@cmdcolin)

🐛 Bug Fix​

  • #4920 Fix issue where multi-level linear synteny view would show a blank tracklist (@cmdcolin)
  • #4913 Fix parent aggregation in BigBed files using larger request window (@cmdcolin)
  • #4905 Fix bug where applying genotype clustering twice produces randomness (@cmdcolin)
  • #4897 Fix clustering for certain sets of bigwig files (@cmdcolin)

Committers: 1​

v3.1.0 Release

· 3 min read

This release adds several interesting features including

Sticky view headers and pinned tracks​

We created a new UI where the "view header" stays visible even when scrolling the page.

We also added the ability to "pin" tracks in the linear genome view, which also stay visible when scrolling the page.

Cluster multi-quantitative tracks​

Motivated by the clustering workflow in the Multi-variant viewer, a similar feature was added to the multi-quantitative track

To do this, we sample the data according to the current zoom level (otherwise, it would create very large matrices at per-base resolution) and then provide an R script the user can run to cluster the data

Before clustering image

After clustering Screenshot From 2025-03-11 00-25-06

Simplify some adapter configuration​

The configuration system has a lot of complex substructure, but in this release we created the ability to supply a simplified config, and the results will be auto-determined

For example

{
  "type": "AlignmentsTrack",
  "trackId": "volvox_alignments",
  "name": "volvox-sorted.bam",
  "assemblyNames": ["volvox"],
  "adapter": {
    "type": "BedGraphTabixAdapter",
    "bedGraphGzLocation": {
      "uri": "file.bed.gz"
    },
    "index": {
      "location": {
        "uri": "file.bed.gz.tbi"
      }
    }
  }
}

after

{
  "type": "AlignmentsTrack",
  "trackId": "volvox_alignments",
  "name": "volvox-sorted.bam",
  "assemblyNames": ["volvox"],
  "adapter": {
    "type": "BedGraphTabixAdapter",
    "uri": "file.bed.gz"
  }
}

The CLI tools and other systems will continue to output the verbose version for the time being, but this simplified system will be handy to anyone who does hand-coding of the configuration

Downloads​

To install JBrowse 2 for the web, you can download the link above, or you can use the JBrowse CLI to automatically download the latest version. See the JBrowse web quick start for more details.

yarn run v1.22.22 $ lerna-changelog --silent --silent --next-version 3.1.0

3.1.0 (2025-03-14)​

🚀 Enhancement​

  • Other
    • #4876 Allow inferring default values of adapter urls to simplify config.json (@cmdcolin)
    • #4875 Add sample data from HGSVCv3 to 1000 genomes demo and config_demo (@cmdcolin)
    • #4880 Allow selecting track error messages with click and drag (@cmdcolin)
    • #4882 Allow clustering multi-wiggle tracks on the fly (@cmdcolin)
    • #4874 Use a '012' matrix to improve clustering of genotypes (@cmdcolin)
  • core
    • #4871 Add deletion, insertion, skip, softclip, hardclip to theme (@cmdcolin)
  • app-core, core, product-core

🐛 Bug Fix​

Committers: 2​

v3.0.5 Release

· One min read

Fixes bug with trix searching that has existed since v3.0.0. Thanks to @abjanssen for reporting

Downloads​

To install JBrowse 2 for the web, you can download the link above, or you can use the JBrowse CLI to automatically download the latest version. See the JBrowse web quick start for more details.

yarn run v1.22.22 $ lerna-changelog --silent --silent --next-version 3.0.5

3.0.5 (2025-03-04)​

🐛 Bug Fix​

Committers: 1​

v3.0.4 Release

· 2 min read

New release

Downloads​

To install JBrowse 2 for the web, you can download the link above, or you can use the JBrowse CLI to automatically download the latest version. See the JBrowse web quick start for more details.

yarn run v1.22.22 $ lerna-changelog --silent --silent --next-version 3.0.4

3.0.4 (2025-03-01)​

🚀 Enhancement​

  • Other
  • core
    • #4854 Add status callback for parsing PAF, BLAST tabular, MashMap files (@cmdcolin)

🐛 Bug Fix​

  • core
    • #4862 Fix name indexing failure on jbrowse desktop (@cmdcolin)
    • #4851 Fix NCBI sequence aliases refNameAliases adapter for some sequence_report.tsv files (@cmdcolin)
  • Other
    • #4849 Fix height of Hi-C track, and add adjust-to-height (@cmdcolin)
    • #4845 Fix retrieval of VCF metadata for tooltips in VCF feature details (@cmdcolin)
    • #4842 Fix occasional 1px gap in sequence track (@cmdcolin)

🏠 Internal​

Committers: 1​

v3.0.3 Release

· One min read

Fixes modifications drawing bug in v3.0.2

Downloads​

To install JBrowse 2 for the web, you can download the link above, or you can use the JBrowse CLI to automatically download the latest version. See the JBrowse web quick start for more details.

yarn run v1.22.22 $ lerna-changelog --silent --silent --next-version 3.0.3

Done in 0.44s.

v3.0.2 Release

· 2 min read

This release mostly has small bugfixes and minor improvements

  • Fix for some CRAM files that use "archival" bzip2 codecs
  • Change multi-variant viewer to use a blue color for the "major ALT allele" rather than just ALT===1
  • Use assembly "displayName" in view headers
  • Group transcripts from GTF files into gene level features
  • Improved ?hubURL behavior to load defaultLocation, assembly names
  • Handle random color generation for >10 categories in the multi-variant viewer

Downloads​

To install JBrowse 2 for the web, you can download the link above, or you can use the JBrowse CLI to automatically download the latest version. See the JBrowse web quick start for more details.

yarn run v1.22.22 $ lerna-changelog --silent --silent --next-version 3.0.2

3.0.2 (2025-02-13)​

🚀 Enhancement​

  • core
    • #4834 Update ReExports for @jbrowse/core/BaseFeatureDetails (@cmdcolin)
    • #4818 Improve status updating for BamAdapter, CramAdapter (@cmdcolin)
    • #4811 Improvements to the sesson/config warning dialogs (@cmdcolin)
  • Other
    • #4829 Less gappyness at SVG features block boundaries (@cmdcolin)
    • #4828 Additional ?hubURL compat (@cmdcolin)
    • #4827 Improve status updating for BedTabixAdapter, Gff3TabixAdapter, VcfTabixAdapter, BigBedAdapter (@cmdcolin)
    • #4825 Handle multiple jobs with the same name (@garrettjstevens)
    • #4826 Calculate most frequent non-ref allele for multi-variant renderer (@cmdcolin)
    • #4824 Aggregate transcript level features in GTF based on "gene_name" field (@cmdcolin)
    • #4810 Ensure embedded users get the latest semver @gmod/vcf for multi-variant fixes (@cmdcolin)
  • app-core, core

🐛 Bug Fix​

  • sv-core
    • #4837 Update @gmod/cram for fixed decoding of some CRAM files (@cmdcolin)
  • Other
    • #4808 Fix random palette color generation for multi-variant colors (@cmdcolin)

📝 Documentation​

🏠 Internal​

Committers: 2​

2024 year in review

· 5 min read

Hello all,

I know we are well into 2025 already, but I wanted to review some of the great progress JBrowse made in 2024!

Without further ado, here are some highlights

New multi-sample VCF displays​

To help render population-scale data, we created new "multi-variant" rendering modes. This includes a "matrix" rendering mode that is a dense visualization of variation patterns.

This was an exciting development that has created a flurry of feature development

One particularly interesting feature is the "phased rendering mode", which can split the phased genotypes into separate rows, which effectively shows the individual haplotypes.

The "phased rendering mode" has the unique ability to help spot patterns of inheritance in e.g. trio datasets. You can visually see where a child inherited particular variants from a particular parent

The data in the multi-variant display can be sorted according to sample metadata, which can be imported from a "samples TSV", so users can group and color the samples by e.g. the population code (here showing 1000 genomes data)

Multi-way synteny views​

We created the basis for rendering multi-way synteny views. Currently, this is based on "series of pairwise" comparison rendering, which has certain drawbacks (it is not a true multi-way comparison) but it is a big stepping stone

We intend to create more improvements allowing users to load multi-way comparisons from a single data file, and easier UI workflows for setting up multi-way views in the future

Inversion SVs enhancements for breakpoint split view​

We created new enhancements to the breakpoint split view to allow it to work better in a 'single row' configuration, and to color reads by pair orientation or by changes in direction in a split alignment

Screenshot showing read pairs being connected from a variety of sequencing runs from Illumina, PacBio, and Nanopore from the Genome in a Bottle dataset

Note: There were many other improvements to the breakpoint split view including, importantly, the ability to launch it from non-SV-inspector workflows. Previously, the SV inspector was the only way to really launch a breakpoint split view, but now it can be launched from just clicking the feature details of a read or VCF feature.

New "SV inspector" improvements​

  • Allows opening tracks from the config.json in the SV inspector and spreadsheet
  • Uses @mui/x-data-grid for display of the spreadsheet
  • Makes separate columns for VCF INFO fields
  • Support larger files by storing data in 'volatile' storage that doesn't bog down the session
  • Support DEL, INV, and all types of SVs instead of just TRA and BND type variants in the SV inspector
  • Re-navigates existing breakpoint split view instead of launching a new one each time

Screenshot showing the new SV inspector user interface

New "Recent sessions" menu with autosaves​

We saw users have trouble where they would close their tab containing JBrowse 2 and they would lose their work.

It would be difficult to restore where they were at

To help this situation, we created a new system where sessions are AUTOSAVED! Yes, autosaved! So if you accidentally close a tab or come back the next day, it is easy for you to go back and restore where you were at:

Screenshot showing the ability to access "File->Recent sessions"

  • You can also create "Favorite" sessions to revisit for later
  • Admins can also create "Pre-configured" sessions for their users

If you have any feedback about this feature let us know. I think the discoverability of this feature could use more improvement

New "Group by" functionality for alignments tracks​

We released new "Group by" functionality for the alignments tracks. This functionality is somewhat unique because applying it creates multiple functionally independent subtracks.

The effect of being split into independent subtracks is actually different from e.g. the Group by feature in IGV, because in IGV, only the "read stack" is split into different rows, whereas with JBrowse, each group gets its own read stack and coverage calculation

image8

Example with splitting an alignments track by the HP tag

Apollo 3 beta release​

Apollo 3, the genome annotation editor built on JBrowse 2, went into public beta at the end of 2024!

See release announcement here https://github.com/GMOD/Apollo3/releases/tag/v0.3.0

The Apollo 3 editor can run as a 'collaboration server' (shared annotation editing on a website) or as a standalone JBrowse 2 desktop plugin. It is able to take full advantage of JBrowse 2 capabilities, and users can annotate e.g. inside of a synteny view

image

Feel free to get in touch for more information on the project!

The year ahead​

We look forward to hear from more users. Our office hours outreach effort has been very successful with over 100 one-on-one meetings with users! Feel free to sign up for one at https://jbrowse.org/jb2/contact :)

v3.0.1 Release

· One min read

This release fixes the vanillajs build for the linear genome view the hover genotype feature in v3.0.0

Downloads​

To install JBrowse 2 for the web, you can download the link above, or you can use the JBrowse CLI to automatically download the latest version. See the JBrowse web quick start for more details.

yarn run v1.22.22 $ lerna-changelog --silent --silent --next-version 3.0.1

Done in 0.69s.

v3.0.0 Release

· 7 min read

Hello all!

This release announces v3.0.0 of JBrowse 2. It is not at all a dramatic change like the JBrowse 1 -> JBrowse 2 transition, it's much more incremental, and in fact, you can go on calling it JBrowse 2...as a point of comparison, ggplot2 is on v3.5.1 right now :)

We decided to make a major bump due to some small "breaking" changes that could affect plugin and embedded library users in particular.

But for most users, this release should have nothing but the usual bugfixes and improvements! And there are a lot of these!

Without further ado, here are some highlights!

Improved rendering of phased VCF in multi-variant view​

Phased variants offer a unique opportunity to see which parent particular variants came from. To help this, we added a new "phased" rendering mode to the multi-variant renderings, that takes phased VCF and renders each phase as an individual row. This allows users to effectively see the entire row as a haplotype, and can be used to help visualize trio VCF to find parent of origin

Screenshot From 2025-01-25 15-03-21 Screenshot showing the "phased mode" for a trio VCF

image

Screenshot showing how to access the phased rendering mode

Rendering of phase-sets​

This could be more or less uncommon depending on your work, but for incompletely phased variant calls you will see "phased sets" (PS tag in genotypes). We color each set a random color

image

Screenshot showing phase set rendering of a phased VCF. This is nearly completely phased so just showing green, but each phase set would get a unique color. The black entries are unphased variants (which could be filtered out, but this is not done currently)

Rendering polyploids in the multi-variant view​

image

Screenshot showing the "polyploid" rendering of the multi-variant display.

Polyploid variant calls can look like 0/2/./1 indicating one match to the reference (0), two alts (1,2), and one missing call (.).

  • Yellow indicates missingness
  • Grey indicates reference
  • Blue indicates the ALT===1
  • Red is ALT!==1

Each of these is a spectrum that increases in darkness, so darker blue indicates more ALT===1, darker red indicates more ALT!==1, darker yellow is more uncalled, etc.

Adding sample metadata to VCF from an external TSV​

image

Screenshot showing multi-sample matrix display, with the population colors coming from a "sample TSV" file. The below config shows the samplesTsvLocation config, the first column should match the sample name in the VCF

Example TSV

name	population
HG01879	ACB
HG01880	ACB
HG01881	ACB
HG01882	ACB
HG01883	ACB
HG01884	ACB

Config

{
  "type": "VariantTrack",
  "trackId": "1kGP_high_coverage_Illumina.chr1.filtered.SNV_INDEL_SV_phased_panel.vcf",
  "name": "1kGP_high_coverage_Illumina.chr1.filtered.SNV_INDEL_SV_phased_panel.vcf",
  "assemblyNames": ["hg19"],
  "adapter": {
    "type": "VcfTabixAdapter",
    "vcfGzLocation": {
      "uri": "https://ftp-trace.ncbi.nlm.nih.gov/1000genomes/ftp/release/20130502/ALL.chr1.phase3_shapeit2_mvncall_integrated_v5a.20130502.genotypes.vcf.gz"
    },
    "index": {
      "location": {
        "uri": "https://ftp-trace.ncbi.nlm.nih.gov/1000genomes/ftp/release/20130502/ALL.chr1.phase3_shapeit2_mvncall_integrated_v5a.20130502.genotypes.vcf.gz.tbi"
      }
    },
    "samplesTsvLocation": {
      "uri": "https://jbrowse.org/genomes/hg19/1000g.sorted.csv.gz"
    }
  }
}

Improved UI for opening synteny tracks​

It has always been challenging for end users to open synteny tracks

To help with this, we now support adding synteny tracks via the default "Add track" workflow, and make it easier to specify the query and target assembly names. We also added a similar UI change to the "Linear synteny view" import form

image

Screenshot showing the default add track workflow with new UI for choosing target and query assemblies for PAF tracks

image

Screenshot showing the add track workflow in the "Linear synteny view" import form, also showing new UI for choosing query and target assmblies

New "turbo zoom" buttons​

In working with users, we noticed people having to click the zoom buttons or fiddle with the slider a lot. To aid this, we added a small dropdown for quickly doing 10x, 50x, and 100x zooms

image

For developers: some "breaking" changes​

  1. We changed the "filehandle" type used in data adapters. We now use generic-filehandle2 which is simpler and does not require a node.js polyfill on the web
  2. Upgrading React 18 -> React 19 in our main webapp. This drops support for React 17, and may have some other implications

If you are a plugin or embedded user, and experience problems with upgrading, let us know!

Downloads​

To install JBrowse 2 for the web, you can download the link above, or you can use the JBrowse CLI to automatically download the latest version. See the JBrowse web quick start for more details.

yarn run v1.22.22 $ lerna-changelog --silent --silent --next-version 3.0.0

3.0.0 (2025-01-29)​

💥 Breaking Change​

  • core
  • __mocks__, app-core, core, embedded-core, product-core, sv-core

🚀 Enhancement​

  • Other
  • core, web-core
  • core, product-core
    • #4798 Add more track metadata to UCSC assembly hub loading (@cmdcolin)
  • core
    • #4795 Improved rendering of phased VCF and polyploid VCF in multi-variant view (@cmdcolin)
    • #4792 Allow uploading a sampleTsv file with per-sample metadata for VCF files (@cmdcolin)
    • #4786 Allow opening "synteny tracks" from the default Add track workflow (@cmdcolin)
    • #4785 Allow opening BedGraph et al in the default Add track workflow (@cmdcolin)
    • #4784 Improve UI for choosing query and target assemblies in the "New track" workflow for synteny tracks (@cmdcolin)
    • #4767 Improve load time when multiple tracks use same data file (@cmdcolin)
    • #4750 Add some more lazy loading for smaller bundle size (@cmdcolin)
  • sv-core
    • #4753 Allow launching a single level breakpoint split view focused on the breakends (@cmdcolin)
  • web-core
  • app-core, core
    • #4729 Update cram-js to avoid requesting file size for more CORS compatibility (@cmdcolin)

🐛 Bug Fix​

  • Other
    • #4778 Fix GC content calculation consistency across blocks (@cmdcolin)
    • #4773 Fix "Color by CDS" for BED12 features (@cmdcolin)
    • #4771 Fix generating categories for composite tracks in UCSC hubs (@cmdcolin)
    • #4769 Fix facet filter crash (@cmdcolin)
    • #4765 Fix repeat masker track BED interpretation (@cmdcolin)
    • #4764 Fix modifications coverage calculations (@cmdcolin)
    • #4731 Fix naming of subtracks produced by "group by tag" operation (@cmdcolin)
    • #4732 Fix persisting alignments "color by" and "filter by" settings in snapshots/session shares/page reloads (@cmdcolin)
  • core
    • #4776 Fix maximum recursion error in linear synteny view import form (@cmdcolin)
    • #4768 Fix navigation via CIGAR string alignment in synteny view (@cmdcolin)
    • #4766 Fix some bed12 transcripts displaying as bedMethyl features (@cmdcolin)
    • #4727 Fix type confusion where sequence feature details crashes for empty subfeatures (@cmdcolin)
    • #4725 Avoid error on closing draggable dialog and a couple more refactors (@cmdcolin)

🏠 Internal​

Committers: 2​