Configurable Y offset for features relative to track label

[dmgood11]

We like having track labels displayed, but could an option exist to offset the label from the features in the track? We often have very deep tracks so the additional depth from having the label Y-offset from the topmost feature is a very small price to pay compared to not being able to read a feature label or click on a feature without scrolling over horizontally to clear the track label.

[cmdcolin]

Possible dupe https://github.com/GMOD/jbrowse/issues/490

Hiding track labels is one way this has been "solved" with HideTrackLabels but I agree it would be better to just have at least the option to move the features out of the way (kind of what I wanted to say on that issue with "reconfiguring the UI")

[dmgood11]

Yes, it's a dupe (my search for similars wasn't extensive enough to find #490). But I haven't come across any of our end users who consider it "solved" by clicking the HideTrackLabels button. IMO it's a significant UX issue.

[cmdcolin]

I guess just conceptually, it could either be solved where every track is given an offset, or each track type could implement an offset, or maybe both. I am not sure which would be easier.

The calculation of track heights is pretty fundamental to how the genome browser works also so it might have to dig into some internals

[scottcain]

This is related to an issue raised at WormBase: https://github.com/WormBase/website/issues/5900 A plugin to allow the tracklabel to be "raised up" relative to the track so that it can't obscure features in the track would be great.

[enuggetry]

Add config feature to move track labels/menus above the features so as not to obscure the features. Adds a little more spacing between tracks and between the first track and the genome view. Activated with trackLabels: "no-block" in trackList.json.

trackList.json
{
   tracks: [
      ...
   ],
   ...
   "dataset_id" : "volvox",
   "trackLabels":"no-block",
   "formatVersion" : 1
}

[rbuels]

Sorry about the delay.

Have you tried setting a larger trackPadding @dmgood11? Something like trackPadding: 70 in the configuration?

[cmdcolin]

I think the new "no-block" option could be applicable here. The trackLabels: no-block is a nice feature, and should be made easy for users to find. Here is how it looks with it on

With view.trackPadding=70 it doesn't move the track label out of the way

Note that the "no-block" feature is sort of awkward to enable. Why not just be a true/false instead of set to specific string? Secondly, no-block works with trackPadding and sets it manually to 35. Kind of hardcoded

[rbuels]

@cmdcolin @dmgood11 So would you guys say this issue is satisfactorily solved by the no-block configuration value available in the new JBrowse 1.12.4?

[cmdcolin]

I think it works pretty well but my wishlist for the feature would be

a) didn't have hardcoding for tracklabel pixel height in no-block b) that no-block didn't move the tracklabel upwards but rather moved track content downwards c) enable it by default or have a checklist in UI to enable it (biggest concern, since it seems like such a common request from users, and users might want it even if admins don't realize) d) make the config value a boolean instead of string

sort of lengthy and bikesheddy but could close in interest of triaging issues

[rbuels]

Well, the global track label toggle is the primary thing that was meant to address c), don't you think that's enough for now?

The rest of the stuff is the kind of thing to worry about in the 2.x series.

[cmdcolin]

Should be sufficient then :) it is a good workaround

[dmgood11]

Yes looks good to me as well (sorry about the late response, we didn't have the latest release rolled out until today).

-David

On Tue, Feb 20, 2018 at 4:20 PM, Robert Buels notifications@github.com wrote:

Closed #901 https://github.com/GMOD/jbrowse/issues/901.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/GMOD/jbrowse/issues/901#event-1483664884, or mute the thread https://github.com/notifications/unsubscribe-auth/AGUE8xVtL2NR-3sGQ2j6ZuiaaJhytzeXks5tW2FLgaJpZM4N9yi9 .

-- David M. Goodstein, Ph.D. Phytozome and Plant Comparative Analysis Group Lead Joint Genome Institute - U.S. Dept. of Energy http://phytozome.jgi.doe.gov