Call the largest subtree number 1

[corneliusroemer]

When I usher 100 sequences which I've queried with a certain mutation query, I usually don't care about all trees, just the biggest or biggest few.

Unfortunately, it's currently not easy to quickly know which of the many subtrees (often 20 when ushering 100+) are the biggest.

It would be super convenient if subtrees were numbered from biggest to smallest: so subtree number 1 would be the tree with the most sequences from the query, subtree number 2 the one with the second biggest number etc.

By subtree number I mean the right most column:

[AngieHinrichs]

I think I already do this -- please send me a list of sequence IDs (or the fasta if that's easier) if the subtree numbers that appear in the results table are not ordered by number of uploaded samples, with subtree 1 having the most.

[russcd]

You are correct, Angie, AFAIK.

I think Cornelius' point might be that the table is not ultimately sorted by subtree size/membership. Cornelius, would you prefer for subtree 1 and all samples in that subtree to be placed at the top of the table?

On Mon, Nov 7, 2022 at 1:15 PM Angie Hinrichs @.***> wrote:

I think I already do this -- please send me a list of sequence IDs (or the fasta if that's easier) if the subtree numbers that appear in the results table are not ordered by number of uploaded samples, with subtree 1 having the most.

— Reply to this email directly, view it on GitHub https://github.com/yatisht/usher/issues/299#issuecomment-1306202038, or unsubscribe https://github.com/notifications/unsubscribe-auth/ACMZAMJVRS5F2LRKOLUQ7KLWHFWQRANCNFSM6AAAAAARX6YKTE . You are receiving this because you are subscribed to this thread.Message ID: @.***>

-- Russ Corbett-Detig Associate Professor Department of Biomolecular Engineering University of California, Santa Cruz

[corneliusroemer]

I see! You may be right that the numbering is correct - in that case it's two things that could be improved:

• ordering (as @russcd mentioned)
• stating the number of samples in the subtree in the table or in parentheses (see #300)

Thing is, I never know which trees I have already clicked. Having it sorted would be very helpful for that. I could then just go in increasing order rather than having to jump around and look at link color.