Add "nuc" branch label option parallel to "aa"

[corneliusroemer]

Context

Automatic "aa" branch labels are useful. Sometimes it would be nice to show "nuc" branch labels instead, when there are branches without "aa"s in particular.

Description

Add "nuc" branch labels the way we have "aa" branch labels at the moment.

[jameshadfield]

Add "nuc" branch labels the way we have "aa" branch labels at the moment.

The way we do it now is for branch labels to be defined in the JSON, i.e. they're separate from the mutations defined on the branch. This isn't very efficient, but to see what this looks like for your dataset you can (and perhaps should) add these labels via a python script.

I'd be in favor of allowing Auspice to automatically create aa & nuc labels from the mutations, thus forgoing the need for explicit branch labels for mutations.

[corneliusroemer]

I thought the aa are done automatically in a way that nucs aren't? Maybe the automatic part for aa is in some augur command as opposed to auspice, is that what you're saying? There's a difference between manual branch labels like clade and whatnot and the magic "aa" which is always smaller font.

[jameshadfield]

Maybe the automatic part for aa is in some augur command as opposed to auspice, is that what you're saying?

Yes. Here's the relevant code within augur export v2

the magic "aa" which is always smaller font.

Oh, yes you're right:

auspice/src/components/tree/phyloTree/labels.js

Lines 57 to 60 in 40befc6

	const branchLabelSize = (key) => {
	if (key === "aa") return "10px";
	return "14px";
	};

[corneliusroemer]

So it's a combination of special treatment by augur and auspice.

Any reason we should auto-export nucs in augur export? Though I should probably transfer this issue over to augur then...

[jameshadfield]

Any reason we should auto-export nucs in augur export? Though I should probably transfer this issue over to augur then...

I'd prefer to (or someone to) update Auspice to auto-generate these labels without needing to further bloat the JSON with duplicate nuc + aa mutation data. It shouldn't be too hard. In the meantime however adding a "nuc" branch label via a small script will allow you to see these data on the tree. Does that sound reasonable? Happy for others to weigh in here.