dasmoth / dalliance Goto Github PK

Where a track depends on dynamic remapping of data (via hg18ToHg19 or hg19ToHg18) there should be some visual cue to the user to remind them. Maybe an icon with the track label? Or maybe the track label should have a different colour?

Although the paper says the license is BSD, this information is missing on the git repo.

The interface for jumping to certain genome coordinates should be simpler - presently, chromosome, start and end need to be changed in three separate fields / pull-downs.
The tool should accept both "chr3:10000-20000" and "chr\t10000\t20000" (\t = tabs) strings. Motivation: Commonly, researchers deal with lists of genomic loci that are denoted in one of the two formats above (the latter coming also out of a 3-column Excel format. To quickly inspect such a list of disjoint loci, one copy-paste operation per locus is all it should take to get there (e.g. UCSC enables exactly this). Thanks!

After clicking on a menu item, need to click on track label again to reactivate browser. Bug? Would seem more natural to automatically restore selected track after selected menu item is cleared.

Found dalliance and DAS while looking around for genome browsers, and it's definitely the coolest-looking one I've found. However, despite billing itself as a general purpose genome browser, it seems hard-coded to view eukaryotic genomes, the human one in particular. For instance, it seems hard-coded to query a "karyotype" endpoint for features of type "karyotype." I want to visualize bacterial and archaeal genomes, or even just contigs, so there technically is no karyotype.

So is this really a general purpose browser, or is it only for eukaryotes? Am I missing some configuration options that will just let me show arbitrary features from a DSN?

I get an alert showing the error message in the subject when i load biodalliance.org and any of the examples from the site, in addition, the tracks show a red error box:

"Failed request: server may not support CORS"

Peak calling (as well as several other bioinformatics analyses) often involves selecting a threshold of some kind (e.g. FDR, p-value, peak area). It would be useful to have a UI for quickly finding and browsing through a set of regions that contain peaks around a specified threshold value to develop an visual intuition for the resulting FP/FN cases.

When data is at the very start of the track it is overlapped by the track label/name. An option to indent the track so it starts just to the right of the track label would be very useful for this situation.

A temp fix for us was to add an option to set: left: 95px; on the class: *.dalliance .view-holder

Some of the features in 0.6.5 like search by gene,export to PDF/SVG, scrolling by the chromosome location are no longer existing in 0.8. Are they going to be available again in the next release?

• Shouldn't be case-sensitive (except Drosophila...)
• Ambiguous matches should be shown in a pop-up?
• Federate searches across multiple datasources?
• Suggest-as-you-type?

Currently there are hardcoded external assets that go over http, such as:

http://www.biodalliance.org/release-0.10/img/loader.gif
and
http://www.biodalliance.org/release-0.10/css/dalliance-scoped.css

could these please be changed to // to allow http + https support

In Hub test, deselecting Dnase track doesn't remove it. Selecting it again adds the track again. Removing trackhub does not fix problem. Clicking on multiple Dnase tracks on display doesn't remove them either!

Hi there,

It seems like the search function/jump to specific coordinates is not working in firefox. When I click on the chromosome location nothing happens. In Chrome a window appears to type in the gene name or location.
I am using Firefox 5.0 on Ubuntu 10.10.

Pascal

When exporting current view in SVG only half visible window is exported. (Dalliance 0.11)

apparently some browsers dont support the binary data. there is (if i recall) an alert indicating that, but it should further mention
which operating systems + browsers are supported, something simple like:

windows: firefox > 4.0.0, safari > 1.1.1
linux: chrome > 1.0, firefox > 3.0
osx : safari > 1.2

or whatever, i made those up. this way, the user can have some information on how to see the full data.

in Firefox anyway:
1)Open the add/remove a track dialogue by "+" button.
2)do nothing but close it again.
3) try opening it again and it doesn't appear?

When using shift-arrow to reduce the height of a track, at some point the track can get no thinner because of the label, but the height of the feature can continue to be reduced. However, the lower scale label does not move with the scale of the feature and vanishes from sight instead.

Should be able to do both a small chromosome overview (as was implemented in Dalliance 0.7 and earlier), and an overview of all chromosomes.

Does it make sense to share code between these two?

Banding patterns, etc., should probably be fetched from a bigbed file. Example:

      http://www.biodalliance.org/datasets/GRCh37/karyotype.bb

Propagate errors properly when a stylesheet fails to load or parse.

Ideally, give specific reports of stylesheet syntax errors too (although the browsers don't seem to be a lot of help here...)

Cannot find for EYA2 gene in the latest version. Got "no match for 'eya2'" error message when trying to find that gene. On the other hand, 'EYA2' is displayed when manually locate the location to 20:45,500,000..45,850,000.

Would it be something from the source file?

Just noticed that whereas the Human dalliance instance can map between NCBI36 and GRCh37, mouse can't (there are sources on NCBIm36...)

Both genes and transcripts have multiple types (particularly in GENCODE, as defined here at UCSC - http://genome.ucsc.edu/cgi-bin/hgTrackUi?hgsid=172011164&c=chr21&g=wgEncodeGencode).

At present this is not visually distinguished, making it non-obvious to a user that they are looking at a psuedogene rather than a gene at the extreme. Parts of this are general, parts are specific to gencode - maybe its the stylesheet of the DAS source?

I'm seeing the error 'warning registry query failed' quite a bit from biodalliance. Is this a fault or just the registry being busy?

In test.html, tried to attach threshold commands to histogram bw track, but didn't work.

quantLeapThreshold: 40,
forceReduction: -1,

No red line displayed

It would be nice if the browser could report if a feature is missing when an on the fly mapping from one genome build to another is performed. Some kind of image at the old position and a pop up to explain to the user what is missing might be sufficient.

I took a look at the developer notes but there is no concise description as to how you would go about achieving this. Can you please elaborate?

Another question from me.

I displayed a BigBed file in dalliance and in general it works fine. However I do not see the strand and score information in the browser. I was hoping, that the features are coloured according to their score. The bed file also contains a itemRgb column, but its ignored too. Am I doing something wrong, or is this just not yet implemented?
Further I would like to add binary bed files as a default track. Is there a way to add binary files to the "sources" in the HTML file, like the DAS servers?

Thanks,

Pascal

It would be really nice if tracks could be grouped and the groups would be indicated visually with labels. We don't need UI to configure that, we're creating them via JS.

I've created an image how I would like it to work. On it you can see two groups of tracks (for subset 1 and 2)

I'm not sure if this has been discussed before or whether I can't find the button, but to me it would be useful to have a "full screen" button at the BioDalliance home page. With full screen I mean that the viewer takes the entire browsers width.

Keep up the good work!

Most easily seen on gene tracks when looking at a zoomed-in view.

Can we reproduce the sticky feature labels from Dalliance 0.7 while maintaining the performance of the canvas renderer?

To enable viewing of genes/exons in the forward order (5'->3') if present on the minus strand.

With reference to google groups discussion (https://groups.google.com/forum/#!topic/biodalliance-dev/OjYwCAk2XFo).

Kind regards,

Pär

as of two/three commits ago, the mRNA tracks are no longer being displayed. I have rolled back to a much earlier version of dalliance to get them to display.

I will roll forward incrementally to find out the point at which the issue started and update the ticket.

https://geefu.oadb.tsl.ac.uk/experiments/32

Feature request (or there might be a way to do this that I missed): I'd like to be able to specify information for the karyoscape without a DAS server.

The karyoscape seems to be the only essential part of the browser that relies on DAS. It would be very nice to be able to serve a Dalliance browser entirely from Apache.

By the way, I have to say I am enormously impressed by this project, especially the handling of binary formats in javascript.

Looking at CTA-85E5.7 - chr22: 30500000. There are two entries displayed when transcript box is clicked. When click on top one it has an ENSG label; when click on bottom one it has an ENST label. I think this is a bug around the displaying genes and transcripts...

Feature request:
It would be helpful to have a way to show the translation of a gene in the browser, for example when zoomed in to a gene far enough.

In Firefox 5 when trying to change the region by clicking on the displayed range e.g Chr22:111111231..1313123131 get the following error in firebug:

NUM_REGEXP is not a function
error source line: [Break On This Error] "",q=a;for(j=0;j<h.length;++j){var m=h...n b}var STRICT_NUM_REGEXP=/^[0-9]+$/;

Line 303 of compiled-dalliance.js

after getting the hg18 xenoRefGene table from UCSC downloaded as an extended bed (12) file, i do the following

wget http://hgdownload.cse.ucsc.edu/admin/exe/linux.x86_64/bedToBigBed
chmod +x bedToBigBed

wget  http://hgdownload.cse.ucsc.edu/admin/exe/linux.x86_64/fetchChromSizes
chmod +x fetchChromSizes

./fetchChromSizes hg18 > hg18.chrom.sizes
sort -k1,1 -k2,2n $BED > $BED.sorted
./bedToBigBed $BED.sorted hg18.chrom.sizes $BED.bb

to get a bigBed format.
then i add:

{name:                 'xenoRef',
bwgURI:               '/dalliance/xenoRefGene.gene.hg18.bed.bb',
collapseSuperGroups:  true}

to the test.html
it appears to be loading features (the track has a height greater than the minimum and if i zoom out far, it gives the message about > 25 features), but they are not drawn.

when i wget one of the .bb files from the example test.html and then update test.html it to point to that local .bb, everything works fine. am i doing something wrong in the creation of the bigbed?

Is there away or would it be possible to provide a mechanism whereby a user could highlight a range on the browser then export features found within that range on specific tracks ?

I have the browser embedded in our app and was hoping to be able to 'feed' any features of interest, identified during a user session back to the host application for further processing.

Thank you

the mixture of tabs and spaces make it more difficult to dig into the source and either see what's going on or contribute.

Ideally as a URI that can be e-mailed/tweeted/whatever to other users.

Hi,

With the latest 0.12.0pre2 I am unable to use a TwoBit as reference genome.

The following error appears: Uncaught ReferenceError: readIntBE is not defined

which I would have expected at line 15, after var readInt = bin.readInt;

Its currently a really thick line and dominates the display - I think it should be a hairline

As can be seen on the picture, scales appear on the far left instead of in the middle

Hi,
Are there any pointers regarding how to use the API to add or remove tracks to a browser session that you can share ?

thxs

Been trying to use the 7pre2 release to take advantage of the color formatting. For very small datasets it works fine but for larger data it appears dalliance is trying to slurp in the whole file into memory. It fails to render and usually have to kill the browser. Seems worse w/ Chrome.

Anyone experience the same prob ? have work around ?

Thanks
B

Just update to v0.12 but I got following js error when trying to add DAS custom data:

Uncaught ReferenceError: DASSegment is not defined dalliance-compiled.js:460
pa dalliance-compiled.js:460
H dalliance-compiled.js:440
(anonymous function) dalliance-compiled.js:460

No track being added. Any advice?

When Browser javascript function is invoked it adds a listener to the page onload event. For standard alone pages this is okay but for apps where the browser is loaded as a plugin, the page load event may not fire.

Just a suggestion, but it maybe better to leave the invocation of the browser to the caller and not assume which event they may wish to use to trigger the actual rendering of the browser.

On the same topic, if the browser is deployed within a div on a larger page it appears you have to flush localStorage otherwise the browser has a 'fit' if you try to refresh/reload it insitu. ( admittedly some of this behavior may be the result of the application I'm trying to embed the browser)

p.s.

Would like to commend you on a great piece of work !