update raxml-ng to 2.0.0 + UI refactor by amkozlov · Pull Request #7829 · galaxyproject/tools-iuc

amkozlov · 2026-03-25T14:02:02Z

FOR CONTRIBUTOR:

I have read the CONTRIBUTING.md document and this tool is appropriate for the tools-iuc repo.
License permits unrestricted use (educational + commercial)
This PR adds a new tool or tool collection
This PR updates an existing tool or tool collection
This PR does something else (explain below)

bgruening · 2026-03-25T14:13:02Z

-            </param>
-            <param name="infile" argument="--msa" type="data" format="fasta,phylip" label="Source file with aligned sequences"
-                   help="At least four aligned genomes are needed for RAxML-NG." />
+    <macros>


can you maybe move that into a separate file, that makes it more comprehensible / reviewable I think.

Thanks @amkozlov

sure, like this?

bernt-matthias · 2026-03-26T07:08:36Z

+    --model '$model.auto_seq_type'
+  #elif $model.model_type == 'single_gui':
+    #if $model.seq_type == 'multistate':
+      #set model_matrix = '""'.join(["MULTI", str($model.single_model.multi_state_count),"_",str($model.single_model.subst_matrix)])


I have difficulty to understand what '""'.join(...) is doing..

for multi-state models, model name contains the number of states, eg. MULTI8_GTR for 8 states.

the might be a more elegant way to build this string, but IIRC + didn't work so I used join() instead

But this will produce a string with lots of additional " characters in it (two quote characters in between each array element will be added), e.g. '""'.join(["MULTI", "8", "_", "COUNT"]) = 'MULTI""8""_""COUNT'.

What you probably want is ''.join(["MULTI", str($model.single_model.multi_state_count),"_",str($model.single_model.subst_matrix)]).

But also "MULTI" + str($model.single_model.multi_state_count) + "_" + str($model.single_model.subst_matrix) should work.

ok I changed it as you suggested and it works - thanks for the hint!

bernt-matthias

We could definitely need a better test coverage for this tool? In particular for such a larger restructuring?

bernt-matthias · 2026-03-26T13:06:08Z

    </xrefs>
    <requirements>
-        <requirement type="package" version="@TOOL_VERSION@">raxml-ng</requirement>
+    <requirement type="package" version="@TOOL_VERSION@">raxml-ng</requirement>


Please restore the old correct indentation.

bernt-matthias · 2026-03-26T13:13:17Z

       </data>
       <data format="nhx" name="startTree" from_work_dir="galaxy.raxml.startTree" label="${tool.name} on ${on_string}: Starting trees">
-          <filter>general_opts['command'] in ["--search", "--search1", "--all", "--start"] </filter>
+           <expand macro="m_raxmlng_nofiles_filter"/>


the filetype token needs to be used here, or? Maybe also in other outputs?

I guess I will rather remove this token and move the respective logic out of the macros,

I'm trying to do some (pretty convoluted) output filtering, consider this example:

<data format="nhx" name="startTree" from_work_dir="galaxy.raxml.startTree" label="${tool.name} on ${on_string}: Starting trees"> <filter> general_opts['cmdtype']['command'] == "--start" or (general_opts['cmdtype']['command'] in ["--search", "--fast", "--all", "--allfast"] and output_opts['keep_interim_files'] and "startTree" in output_opts['keep_interim_files']) </filter> </data>

Intended behavior:

startTree file is primary result for one command (--start), but rarely used intermediate results for others

so for those other commands, startTree file will be ignored by default - unless user sets a respective checkbox (in output_opts['keep_interim_files']) to indicate that they do want to see this file in the output

btw I noticed a weird thing:

if no checkboxes are set and output_opts['keep_interim_files'] is empty (undefined?), then "startTree" in output_opts['keep_interim_files']evaluates to true.

is this intended behavior or was I hallucinating like with join()? :)

"startTree" in output_opts['keep_interim_files'] will trigger an Exception (TypeError) if output_opts['keep_interim_files'] unchecked (then it is None).

If a filter triggers an exception the file will be included. Therefore you need to check for

output_opts['keep_interim_files'] and "startTree" in output_opts['keep_interim_files']

bernt-matthias · 2026-03-26T13:15:17Z

+           <expand macro="m_raxmlng_nofiles_filter"/>
+           <filter>general_opts['cmdtype']['command'] in ["--all", "--support", "--bsref"] and "fbp" in general_opts['cmdtype']['bs_metric']</filter>
+       </data>
+       <data format="nhx" name="supportTreeTBE" from_work_dir="galaxy.raxml.supportTBE" label="${tool.name} on ${on_string}: ML Tree with branch support values (TBE)">


Please cover new outputs in tests?

Sure, will do.
I have a related question:

I have some output files which are optional and depend on the input file content (not input parameters). E.g. a raxml.collapsedBestTree is only created when final tree has short branches that has been collapsed.
How do I define such outputs in XML? Basically, I want the following:

if the file exists, it should appear in history

if the file doesn't exist, it's not an error but there should be no respective history entry

I think this is not possible. If its 1 or more files this might be an option: https://planemo.readthedocs.io/en/latest/writing_advanced.html#examples

Alternatively but such files in a collection, i.e. all such optional outputs in one collection. Then users can extract them from the collection by the name.

ok thanks, I'll look into it!

amkozlov · 2026-03-27T14:35:01Z

We could definitely need a better test coverage for this tool? In particular for such a larger restructuring?

I added some more tests, please have a look.

bgruening · 2026-04-07T19:41:15Z

+    <token name="@TOOL_VERSION@">2.0.0</token>
+    <token name="@VERSION_SUFFIX@">0</token>
+    <xml name="m_raxmlng_msa">   
+        <param name="infile" argument="--msa" type="data" format="fasta,phylip" label="Input file with aligned sequences (FASTA/PHYLIP)"


You could add a validator here and ask for fasta and phylip files with sequences >= 4

https://github.com/galaxyproject/galaxy/blob/4220f5b4240014fd4a6aa8e69bace06551f23ff6/lib/galaxy/datatypes/phylip.py#L38

done, thanks for the hint!

update raxml-ng to 2.0.0 + UI refactor

62a7bc3

bgruening reviewed Mar 25, 2026

View reviewed changes

amkozlov added 5 commits March 25, 2026 15:29

raxmlng: move macros to a separate file

c61398d

raxmlng: fix lint errors

27498fa

raxmlng: fix linter warnings

fbca982

raxmlng: trye to fix unit tests

f843610

raxmlng: try to fix unit tests again

c558e7f

bernt-matthias reviewed Mar 26, 2026

View reviewed changes

amkozlov added 9 commits March 26, 2026 16:46

raxmlng: use + instead of join() for string concatenation

578464e

raxmlng: nofiles_filter

0dd417f

raxmlng: add optional output: bestTreeCollapsed

7b90be7

raxmlng: add a few more tests

15df681

raxmlng: fix tests

4187be4

raxmlng: fix tests 2

5b56b07

raxmlng: fix tests 3

e366f1a

raxmlng: more tests

81fc32a

raxmlng: fixed failed test

cdabdc3

bernt-matthias mentioned this pull request Mar 30, 2026

Updating tools/raxmlng from version 1.2.2 to 2.0.0 #7846

Closed

bgruening reviewed Apr 7, 2026

View reviewed changes

amkozlov added 2 commits April 13, 2026 17:37

raxmlng: add sequence number validator for the input MSA

d78677d

update raxml-ng website URL

1baec10

Conversation

amkozlov commented Mar 25, 2026

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

bernt-matthias left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

amkozlov commented Mar 27, 2026

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants