User guide

Navigating the UI

Linear genome view usage

To start a linear genome view, use the menu bar

File->Add->Linear genome view

Using the location search box

• Use the search box in the LGV
• Enter syntax chr1:1-100 or chr1:1..100
• You can also specify an assembly name with the locstring {hg19}chr1:1-100

You can also use the search box to search by gene name (if it is configured)

Figure: Search for gene names or other features via the location search box.

In order to enable name searching, you or the admin on the instance will need to create a "text index". See the quickstart guide for more info.

Scrolling

Mouse wheel can scroll side to side, as well as click and drag. The pan buttons also exist in the header of the linear genome view

Zooming

The zoom buttons exist in the header of the linear genome view, and there is also a slider bar to zoom in and out.

Note: You can also hold the "Ctrl" key and use your mousewheel or trackpad to scroll and this will zoom in and out

Re-ordering tracks

There is a drag handle on the track labels indicating by the six dots, clicking and dragging on this part of the track label can reorder tracks

Adding tracks

To add a new track or connection, you can use the menu bar in the app to open the form for adding a track:

File->Open Track

Figure: After opening the menu item for 'Open new track' a drawer widget for the 'Add track form' will appear

Note: There is also a circular "+" button inside the track selector menu that can also be used to access the "Add track" form.

Figure: The orange plus icon in the bottom right of a tracklist can also be used to launch the 'Add track form'

In the "Add track" form, you can provide a URL to a file to load, or you can also open files from your local machine. In some cases, you need to provide an index (bigwig files for example have no index, but BAM/CRAM or tabix filetypes like VCF/GFF/BED tabix do). In some cases we can automatically infer the index e.g. if you provide a URL for a BAM and the index filename is bamfilename +'.bai' but you may need to manually supply it in some cases (index inference can't be done with files from your local machine)

The following file formats are supported

• Tabixed VCF
• Tabixed BED
• Tabixed GFF
• BAM
• CRAM
• BigWig
• BigBed
• .hic file (Juicebox)
• PAF

For tabix files, TBI or CSI indexes are allowed. CSI or BAI is allowed for BAM. Only CRAI is allowed for CRAM. The index will be inferred for BAI or TBI files as filename+'.bai' for example, but if it is different than this, make sure to specify the index file explicitly.

Note: If you are an administrator, you can add tracks with the command line or with the admin server add-track or admin-server guide

Sharing sessions

On JBrowse Web, the main menu bar has a "Share" button to enable users to share their sessions with other people. The share button generates a URL that can be sent to other users.

Note: Share sessions is not available on JBrowse Desktop.

Also note: you can't copy the literal URL in your address bar and send it to other users, you must use the "Share" button to do that, but you can copy and paste the URL in your URL bar between different tabs in your local browser though

Figure: The session share dialog, which gives you a short URL to share your session with other users. It is important to use these URLs generated here rather than simply copying and pasting your URL to other users

The session URL will contain

• what views are on the screen, and settings for the views (e.g. track labels overlapping or offset)
• what tracks are in the view
• extra tracks that you added with the "Add track workflow"
• for the alignments track, the show soft clipping and sort settings on the pileup
• etc

All this stuff gets included in the session

This means you can share links with your custom tracks with other users, without being a JBrowse admin!

Track menu

Users can access track-specific functions by using the track menu, which is accessible from the track selecter itself ("..." icon) or on the track label (vertical "..."). Some functions are only available when the track is open e.g. from the track label, but more basic options like "About track" are available from the track menu on the track selector.

Figure: Screenshot showing how to open the track menu (both in the track selector area and in the track label area of the linear genome view), and an example of a VCF track with it's track menu open

About track dialog

Using the track menu as described above, you can access the "About track" dialog

Figure: Screenshot of the 'About track' dialog for a CRAM file, showing the full CRAM file header and config info. Having the full header of a BAM/CRAM file available is helpful to easily check what genome it was aligned to for example.

Editing track configs

As a non-admin user, in order to edit a track config, you have to make a copy of the track. This will copy it to your "session tracks", which you can edit freely

Figure: Screenshot showing the procedure to copy the track before being able to edit the settings

Rubberband selection

The scale bars accept a click-and-drag action to select a region. Rubberband selection can be performed on both the main (lower) and overview (upper) scale bars.

Figure: Screenshot of rubberbanding both the main and overview scalebars. The main scalebar produces extra options on selection, e.g. Zoom to region, Get sequence, etc.

Track label positioning

Track labels can be positioned on their own row or overlapping the data to save vertical screen space. They can also be hidden. This is done by clicking on the hamburger menu for a specific view.

Figure: Example of using the overlap and offset track label positioning options

Horizontally flip

The view can be horizontally flipped, or reverse complemented, to make the coordinates go from right to left instead of left to right

We use triangles pointing in the direction of the orientation in the overview bar to help indicate whether the app is horizontally flipped or not

Here is an example of before and after horizontally flipping the view

Figure: Before and after horizontally flipping

Sequence track

The sequence track shows the reference sequence and a three-frame translation. If the view is horizontally flipped, the sequence is "reverse complemented" so the bottom three translation frames go to the top, and the top frames go to the bottom.

Figure: The sequence track, with a positive strand gene for context, shows the start codon on the first exon in the top-three rows of the translation frame. The bottom panel shows the same view but with the view horizontally flipped, and the gene is now shown reversed and the start codon is in the bottom translation frames.

Alignments tracks

Visualizing alignments is an important aspect of genome browsers. This guide will go over the main features of the "Alignments track"

The alignments track is a combination of a pileup and a coverage visualization

Pileup visualization

The pileup is the lower part of the alignments track and shows each of the reads as boxes positioned on the genome.

By default the reads are colored red if they aligned to the forward strand of the reference genome, or blue if they aligned to the reverse strand.

Coverage visualization

The coverage visualization shows the depth-of-coverage of the reads at each position on the genome, and also draws using colored boxes any occurrence of mismatches between the read and the reference genome, so if 50% of the reads had a T instead of the reference A, half the height of the coverage histogram would contain a 'red' box

Figure: Screenshot showing the alignments track, which contains both a coverage view at the top and a pileup view at the bottom

Show soft clipping

If a read contains bases that do not map the the genome properly, they can either be removed from the alignment (hard clipping) or can be included, and not shown by default (soft clipping)

JBrowse 2 also contains an option to "show the soft clipping" that has occurred. This can be valuable to show the signal around a region that contains structural variation or difficult mappability

Figure: Shows what turning on soft-clipping enables for a simulated long-read dataset. There is a simulated structural variant, a deletion, at this position, so the read has bases that map to the other side of the deletion being revealed by this.

Sort by options

The alignments tracks can also be configured to "sort by" a specific attribute for reads that span the center line.

By default the center line is not shown, but by showing it (Go to the view's hamburger menu->Select "Show center line") then you will obtain a better idea of what the "sort by" option is doing

Showing the center line

Here is how to turn on the center line

1. Open the hamburger menu in the top left of the linear genome view
2. Select "Show center line"

Figure: Illustrates before and after turning on the center line. The center line is an indicator that shows what base pair underlies the center of the view. Note that this is used in the 'Sort by' option discussed below; the sort is performed using properties of the feature or even exact base pair underlying the center line

Sorting by base

Sorting by base will re-arrange the pileup so that the reads that have a specific base-pair mutation at the position crossing the center line (which is 1bp wide) will be arranged in a sorted manner. To enable Sort by base

1. Open the track menu for the specific track using the vertical '...' in the track label
2. Select 'Sort by'->'Base pair'

Figure: Illustrating the pileup re-ordering that happens when turning on the 'Sort by'->'Base pair'. The sorting is done by specifically what letter of each read underlies the current center line position (the center line is 1bp wide, so sorted by that exact letter)

Sort, color and filter by tag

We can now also do things like

• Sort by tag
• Filter by tag
• Color by tag

With these features, we can create very expressive views of alignments tracks. For example, in the below step-by-step guide, it shows how to color and sort the reads by the HP tag.

Figure: Step-by-step guide showing how to sort and color by haplotype with the HP tag

Color by modifications/methylation

If you have data that marks DNA/RNA modifications using the MM tag in BAM/CRAM format, then the alignments track can use these to color these. It uses two modes

1. Modifications mode - draws the modifications as they are
2. Methylation mode - draws both unmodified and modifified CpGs (unmodified positions are not indicated by the MM tag and this mode considers the sequence context)

Figure: The track menu can be used to access the settings to color by modifications or methylation

Figure: Screenshot showing the same track in both modifications mode and methylation mode

Figure: After the setting has been enabled you can revisit the dialog box to see the current coloring settings

Color by orientation

JBrowse uses the same color scheme as IGV for coloring by pair orientation. These pair orientations can be used to reveal complex patterns of structural variation

See https://software.broadinstitute.org/software/igv/interpreting_pair_orientations for a good guide on interpreting these pair orientations

Figure: This shows an inverted duplication, the tandem duplication can produce green arrows which have reads pointing in opposite directions e.g. <-- and -->, while blue arrows which can indicate an inversion point in the same direction e.g. --> and -->

BigWig tracks

Visualizing genome signals, whether it is read depth-of-coverage or other signal, can often be done by using BigWig files

Figure: A simple wiggle track with the XY plot renderer

Viewing whole-genome coverage for profiling CNV

The latest jbrowse also allows refining the resolution of BigWig tracks, and viewing whole genome coverage. This allows us to get detailed global views of CNV for example from whole-genome coverage profiling

Here is a short picture guide to setup a whole-genome view of a BigWig for CNV coverage visualization

1. Open your BigWig track
2. Go to the view menu and select "Show all assembly regions"
3. Adjust the "Autoscale type" to your liking, the new options for "Local +/- 3sd" allows the autoscaling to avoid outliers
4. Go to the track menu and select "Turn off histogram fill", which then shows only a dot for each point on the graph
5. Go to the track menu and select "Resolution->Finer resolution" a couple times until resolution looks nice

Also note: all tracks have a drag handle on the bottom of it which you can drag down to make the track taller

Figure: A step-by-step guide to view a whole-genome CNV profile of coverage from a BigWig file

Variant tracks

Visualizing variant tracks from the VCF format alongside the original alignment evidence track is a common workflow for validating your results. In JBrowse 2 we can open a variant track and an alignments track as shown below

Figure: Variant track indicating a SNP alongside the alignment track evidence

Variant widget

The variant features have a specialized widget that contains a table indicating all the calls that were made in a multi-sample VCF. Some VCF files, like the 1000 genomes VCF, can contain thousands of samples in a single file. This table can display the details

Figure: Screenshot showing the variant feature sidebar with a filtered by genotype (with alternative allele '1'). Users can also filter by sample name or other attributes

Linear synteny and dotplot views

The dotplot view is a 2D comparative view that can display alignments between different genome assemblies, or even compare a long-read or NGS short-read versus the genome

Opening a dotplot view

Currently the workflow for launching a dotplot is done by navigating in the header bar to the File->Add->Dotplot view

This will let you select the genome assemblies of interest

Then you can also provide a synteny file in the form of PAF via the Add track workflow

Then currently you must configuration edit the PAFAdapter to indicate the two assemblies in the PAFAdapter

Figure: Adding a new dotplot or synteny view via the menubar

Figure: Example of the import form for a dotplot or synteny view. Allows you to select two different assemblies and a PAF file can be supplied via a URL

Figure: Example of a dotplot visualization of the grape vs the peach genome

See the dotplot configuration for more detailed descriptions

Opening a linear synteny view

Use the main menu bar to select

File->Add->Linear synteny view

Figure: Example of the import form for a synteny view allowing you to select two different assemblies and optionally adding a PAF file via a URL

Figure: Example screenshot showing the linear synteny view for grape vs peach

See the linear synteny configuration for more details on manually configuring the synteny view

Opening a synteny view from a dotplot view

We have designed JBrowse 2 to be able to open up a synteny view from a dotplot view. This is enabled by "display modes" so that the same track can be displayed in different contexts.

Here is a short demo that shows opening a synteny view from a dotplot view selection

Figure: Screenshow showing the 'click and drag' selection over the dotplot view prompts you to open up a linear synteny view from the selected region

Long read vs reference plots

One can also launch a dotplot view that compares a long read to the reference genome by

• Right clicking an alignment
• Select "Dotplot read vs ref" or "Linear read vs ref" in the context menu

Figure: Example of a dotplot of a long read vs the reference genome

Figure: Example of a 'synteny' view of a long read vs the reference genome

Hi-C tracks

Visualizing Hi-C data can be performed with .hic files which are generated by the Juicebox software suite. It uses the hic-straw module developed by the juicebox/igv.js team to visualize it in jbrowse.

Currently configuration options are basic for Hi-C tracks, see configuration for info about configuring Hi-C tracks

Figure: Screenshot of a Hi-C track

SV inspector

The SV inspector is a "workflow" that is designed to help users inspect structural variant calls

Opening the SV inspector

We can start the SV inspector by launching it from the App level menu bar

Figure: The SV inspector can be launched from the main menu bar

This will bring up an "import form" that asks you for your SV evidence. This can be provided opening a file locally or using a URL for files in the following formats:

• VCF or VCF.gz (plain text VCF, or (b)gzipped VCF)
• BEDPE
• STAR-fusion result file

Figure: The import form for getting started with the SV inspector

Example SV inspector workflow

We can start the SV inspector workflow by opening up this file containing translocation events called from a breast cancer cell line SKBR3, based on these published data http://schatz-lab.org/publications/SKBR3/

## Example VCF for use in the SV inspector
https://jbrowse.org/genomes/hg19/skbr3/reads_lr_skbr3.fa_ngmlr-0.2.3_mapped.bam.sniffles1kb_auto_l8_s5_noalt.new.vcf

Copy this URL and paste it into the import form and select hg19

Figure: The SV inspector with the import form and URL pasted

SV inspector results

After loading the user's requested file, you will have a tabular view with each row representing a row of the file you opened, along with a whole-genome overview of the SVs on the right

Figure: The SV inspector with loaded results

Now here is where things can become interesting

We can search and filter the table, with filtering and searching being reflected in the circular view as well.

Figure: The SV inspector with filter applied

Launching breakpoint split view

By clicking on the features in the circular view, or clicking on the triangle drop-down on the leftmost column of the table, we can dynamically launch a new view of the data that is called the "split view" or the "breakpoint split view"

This allows us to inspect the breakpoints of the structural variant, and compare each side to the alignments.

Figure: Screenshot of the 'breakpoint split view' which examines the breakpoints of a structural variant, e.g. an interchromosomal translocation, and connects supporting reads (black splines) and the variant call itself (green thicker line, with feet indicating directionality)

Getting the protein sequence for features

If you have a track with gene or transcript level features, then the feature detail sidebar will automatically stitch together the sequence for that feature. The options include:

• CDS - the coding sequences, spliced together
• Protein - performs protein translation on the CDS, currently assuming the default codon translation table
• cDNA - the CDS plus UTR, or just all exons if a non-coding gene
• Gene w/ introns - the entire gene region sequence with the introns included
• Gene w/ 10bp of introns - the spliced gene sequence with 10bp around the splice sites shown
• Gene w/ 500 up+down stream - the entire gene region with 500bp upstream and downstream (shown in light red)
• Gene w/ 500 up+down stream + 10bp of introns - the spliced gene sequence with 10bp around the splice sites shown and the up/down stream shown

Some of the params such as 500bp and 10bp are arbitrarily chosen, if you are interested in adjusting these parameters let us know

Figure: The sequence for the upstream and downstream, exons, and intron sequences shown in the feature details

Using the plugin store

Users can add plugins to their session using the in-app plugin store. The plugin will be added to your "session" which can be shared with the share button (or if you are an admin running the admin-server, then it will be added to the config file).

This can add extra functions or tracks or many other interesting features. For example, if you add the CIVIC plugin, it will automatically add a track that contains the CIVIC cancer gene annotations to hg19.

Note that not all plugins are directly useful from being added (sometimes it needs extra work on the part of the plugin developer to make them useful in the GUI, some plugins require hand editing of configuration files).

Figure: Screenshot showing the plugin store inside the app