search

tskit-dev / what-is-an-arg-paper

Manuscript and code for the "What is an ARG?" paper
1 stars 8 forks source link

Compress Appendix D #426

Closed jeromekelleher closed 9 months ago

jeromekelleher commented 9 months ago

I compressed appendix D down to the two main points (I think) it was trying to make. Paragraph 1 is about tagging events, and paragraph two is about the reality of polytomies and mutational information required to distinguish them.

What I deleted (I believe) was redundant either because the information was already in the figure caption, or unecesseary/inaccurate argumentation.

Polytomies are not enabled by the gARG view - there is absolutely no reason you couldn't have them in an eARG. Statements about approximations have already been made elsewhere.

jeromekelleher commented 9 months ago

@hyanwong can I get your thoughts here?

hyanwong commented 9 months ago

@hyanwong can I get your thoughts here?

Mostly looks good to me, thanks for shortening.

My only comment is the last sentence: I think it is useful to remind people here that the reason we are talking about this is because of gARGs. I think in the new version we don't mention gARGs at all, right? But the point is that a polytomy in an inferred ARG is implicitly representing a summary of multiple events, so polytomies fit naturally into the "deliberately imprecise" gARG framework we have been talking about in the rest of the paper. (True, you could argue that eARGs could be tweaked to represent simulated polytomies, but I think we don't even need to go there: we just say that the situation in individual D_10 is a natural fit for gARGs)

jeromekelleher commented 9 months ago

We shouldn't be arguing for anything one way or another here, it's an informational appendix intended to illustrate fundamental points about biology. Putting in stuff about gARGs and eARGs just obfuscates this. We make the point about using polytomies to represent uncertainty in the Discussion and elsewhere.

hyanwong commented 9 months ago

I still think it is worth mentioning gARGs here, but maybe we will have to agree to disagree.

I could read the appendix as it stands and not realise that it is is any way relevant to the gARG encoding. That seems wrong to me. I'm not saying we make an argument here, just that we point out a link to gARGs.

hyanwong commented 9 months ago

Did you see my comment above about "gametes" in the main text? I don't think we do mention anywhere that Hudson (and others) define ARGs in terms of gametes. We probably should somewhere, right?

jeromekelleher commented 9 months ago

I could read the appendix as it stands and not realise that it is is any way relevant to the gARG encoding. That seems wrong to me. I'm not saying we make an argument here, just that we point out a link to gARGs.

None of the other appendices are about gARGs. They're not intended as part of the central narrative of the paper. We make the points that you discuss repeatedly, so let's leave it there.

jeromekelleher commented 9 months ago

Did you see my comment above about "gametes" in the main text? I don't think we do mention anywhere that Hudson (and others) define ARGs in terms of gametes. We probably should somewhere, right?

I'm looking at that on overleaf

hyanwong commented 9 months ago

I could read the appendix as it stands and not realise that it is is any way relevant to the gARG encoding. That seems wrong to me. I'm not saying we make an argument here, just that we point out a link to gARGs.

None of the other appendices are about gARGs. They're not intended as part of the central narrative of the paper. We make the points that you discuss repeatedly, so let's leave it there.

My inclination, if there is disagreement, is to leave stuff out, so you are probably right here!