You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hello Jennifer!
I’m glad that you plan to improve ghost-tree!
I would like to add another enhancement request to the list.
Currently, for each genus ghost-tree takes the first sequence for foundation tree building. However, it would be great to select the “best” sequence (e.g., the longest one with the least number of ambiguous nucleotides).
The same principal may apply for the other taxonomy levels (e.g., order for LSU as in #60).
With best regards,
Vladimir
The text was updated successfully, but these errors were encountered:
Hi @vmikk! Yes, I'm finally working on ghost-tree more! 😄I added a few features and tested on large datasets and made new unit tests, but if you see anything wrong, let me know.
This request is a great idea. I will be out of town for a little while but hopefully I can look into this in a few weeks.
Hello Jennifer!
I’m glad that you plan to improve
ghost-tree
!I would like to add another enhancement request to the list.
Currently, for each genus
ghost-tree
takes the first sequence for foundation tree building. However, it would be great to select the “best” sequence (e.g., the longest one with the least number of ambiguous nucleotides).The same principal may apply for the other taxonomy levels (e.g., order for LSU as in #60).
With best regards,
Vladimir
The text was updated successfully, but these errors were encountered: