Display gene name rather than mRNA name

[jrr-cpt]

As @nathandunn knows, the TAMU instance had some customizations that we lost when we updated. For our phage annotation, we want the final product name to be applied to the gene, not the mRNA feature. However, when we are annotating in Apollo, we have to change the mRNA feature name for the display to show the name we selected. Is there a way to have the name in the navigation center panel display the gene name instead of mRNA name?

I was reading here, and wondered if the section titled "Edit annotations" dialogbox settings is relevant for us to be able to customize?

@MoffMade

[nathandunn]

@jrr-cpt Are you talking about the update from 2.0.6 to 2.6?

Just trying to clarify the affect you are looking for.

The mRNA name is what is shown within the annotation as below. The gene name will be different due to different naming conventions and due to the potential for isoforms:

Are you wanting to us the gene name instead of the mRNA as you only ever have a single isoform?

Ideally we could do this (in the evidence), which is a Canvas Track. I don't think we have that yes in this version of JBrowse for HTMLFeatures, but I'm sure we could do it in JB2.

This might be worth doing a call or pinging on gitter.

[jrr-cpt]

For phage, there is very little RNA splicing, and we generally are focused on the exon (CDS). We will apply the majority of our annotations to gene and CDS features.

If we have the names in the evidence, that doesn't extend to a promoted feature, though, right? This is a major issue for our annotation workflow right now. A call could be very helpful.

[nathandunn]

The names in the evidence are not promoted as they are implied modifications of the underlying evidence. However, you (and several groups) have requested the ability to use the original annotation. This is a major thrust of the Apollo 3 effort.

@jrr-cpt I'll schedule a call offline to see if we can do a workaround.

[nathandunn]

Note: this may go to Apollo 3.

[nathandunn]

Quick question for @jrr-cpt . . I'm working on a fix here: #2516 for this #2515 . . and I was wondering what the appropriate structure would be.

The original evidence is:

gene -> (Shine, exon)

What my "fix" should has converted them into is:

gene -> (mRNA -> (exon, CDS, Shine) )

My understanding is you don't want the mRNA level. Just wondering if there could be another coding type there.

[nathandunn]

that is, the Shine sequence should be a child of the gene (I have a bug fix that is currently breaking this).

[jrr-cpt]

@nathandunn not sure I follow. Technically, biologically, the SD is in the mRNA (not the DNA, which is the gene feature). But like I said we generally ignore the mRNA and want the SD to linked directly to gene. Same with exon (CDS for us, since introns are rare). Does that help?

[nathandunn]

@jrr-cpt That is my understanding as well.

However, Apollo isn't setup to handle exons and CDS's without mRNA (or other transcript type).

If there is biologically an mRNA in there, then we should model that. In terms of visualization, we will talk about that tomorrow.

[nathandunn]

Solution:

• on mRNA or gene details create a copy button to the right of the name field: "Copy to gene name" . . or "Sync"
• only available if different

[nathandunn]

FYI @jrr-cpt / @MoffMade this should be in dev now

[MoffMade]

I'll fetch and update our instance in the morning to check it out. Thank you!

[MoffMade]

Looks like it works, but a couple things I found in our testing so far

• The "Sync Name to gene/transcript" button does not become active/clickable until you change the gene or mRNA name value, even if the values are already currently different.
• The updated value on the target feature (the one you are copying/syncing the name TO) does not appear in the annotation list until the page is refreshed which is un-ideal