-
Notifications
You must be signed in to change notification settings - Fork 55
New issue
Have a question about this project? # for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “#”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? # to your account
Add pyani tree
subcommand
#187
Comments
I was just looking to see if this was implemented, as I wanted to extract a Newick format tree from pyANI to compare to another clustering method, e.g. visually with https://phylo.io/ https://doi.org/10.1093/molbev/msw080 |
In the interim, it ought to be straightforward to produce a dendrogram in |
Yep, I will try the R code snippet on the linked issue for this. |
@baileythegreen is working on this, just now. There should be a |
I have initially implemented this as an option within The If you wanted to try this now, from the pyani plot -o <output_file> --run_id <run_id> --dbpath <database_file> -l <log_file> --tree but be warned it will generate 16 plots and 10 Newick files. It also adds some new dependencies. |
Newick file output has been modified so there is only one file, with the 'name' of the tree inside a comment. Thus far, I know this format works both with libraries I have tested in Python, and with the https://phylo.io site Peter linked.
|
Can confirm the format works with |
I wrongly assumed this was already on master, is https://github.com/widdowquinn/pyani/commits/tree_186 the latest version of |
It is. Right now it plots trees as part of the plotting subcommand, and does not offer much customisation. I am in the process of refining this, and also creating a separate subcommand that allows more customisation. All of that is to be done on the |
There isn't a (draft) PR for Also I seem to be missing some graphical dependency stuff... getting these warnings multiple times (despite not warning to actually use the display).
The above seems not to write a tree (i.e. something aborts after the warnings). It looks like right now the branch only adds the tree to the heatmap code in the seaborn plotting |
Hi Peter, There was not a draft PR for this; but I've made one here, to make it easier for you to give feedback/comments. The I've responded to your comment on the last commit here; that's an issue that I need to solve; the test suite on my computer seems happy with the current version, but CircleCI here is failing at this point. I'm testing a potential solution right now, but may also discuss this with Leighton tomorrow.
I will look into this; it seems odd. |
Thank you! |
Summary:
Add Newick tree output to
pyani
.Description:
In #186 a question was raised about generating trees from
pyani
directly. At the moment this isn't implemented, but could be done fairly readily. One API implementation for writing might be:or for graphical output:
Current Output:
Not implemented
pyani Version:
v0.3+
The text was updated successfully, but these errors were encountered: