Latest_version_classy.log,Old_version_classify.log
# taxprofiler
l
Latest_version_classy.log,Old_version_classify.log
@James Fellows Yates I reverted
ganon
to the old module with version 1.5.1 and received the same report as the one attached above (Latest_version_classy.log). After discussing this with @Sofia Stamouli it appears that the Old_version_classify.log was generated by a standalone process rather than through nf-core modules. So I suspect that the log file was not correctly generated via the nf-core modules. But I will test it in the afternoon and back to you before you create a issues on the multiqc repo
j
I'm confused...
the
Latest_version_classy.log
above stil lsays 2.0.0 in the header?
So it says 2.0.0 in the latest run?
l
I am not sure if I understand your question. But what I wanted to say was that the log files from version 2.0.0 and 1.5.1 should be same.
But standalone and nf-core/ganon/classy generate different log files with same version
So the module might not have been properly added.
I will test that later this afternoon
j
the log files from version 2.0.0 and 1.5.1 should be same
Same how so? It may not be, the tool has had some settings tweaks etc
So depends on what you mean here
l
Here is the log file got from 1.5.1 through taxprofiler:
j
And this doesn't get displaye din MultiQC?
l
You are right, multiqc failed to pick up the log file. The same thing you saw in the meeting.
j
dafuq
l
Good morning! I tested
ganon
using the standalone command and got the same log file as taxprofiler. But @Sofia Stamouli mentioned that the
Old_version_classify.log
file attached above was downloaded through Git Actions by adding these lines in the
ci.yml
file. This makes sense to me. So I will update
ganon
to the latest module but won’t include it in the MultiQC report for this taxprofiler release. What do you think @James Fellows Yates?
😢 1
j
Fair enough
I'm very confused what's going on though
l
I could change the search pattern to test mutliqc ganon module if you don’t have time
j
(I can't look more deeply into this today, Teo's fever got worse over night so work is out hte window)
You could try, yes, that's not a bad idea...
l
Oh no poor Teo
Have you visited the doctor to check if he got any infections from kindergarten?
j
When they are this age, unless the fever gets really high (and no other symptoms), they just say monitor, they normally get better by themselves as they've got a pretty strong immune system
l
I see. Perhaps you could use some physical ways to lower the temperature, for instance cold tower. But I am sure you know better than me.
j
Yes, we've been doing that, it's a good idea 🙂 He's in a pretty good mood this morning even if very tired and still has a temperatuer, which is still a good sign
l
Ok then I will not take your time and will update you in the PR
❤️ 1
s
Hope Teo get better soon 😊
l
I have commented in the multiqc issue regarding ganon https://github.com/MultiQC/MultiQC/issues/2805
j
Thanks for the comment but I'm really not understanding
The GHA version is the nf-core module version..
And from my (maybe poor) understanding of the python code the prefix shouldnt make a difference, or if it's missing should cause the module to error not just not get picked up
And iirc ganon doesn't produce a lot file, we are just redirecting STDOUt (or am I remembering badly)
l
I haven’t replicate the log file through git action. So I am not 100% sure. If I understand right that’s how Sofia got the that log file.
j
Ok... I don't think it would make much difference though. The same info is in there...
Maybe I need to sit down and add a bunch of debugging code
l
I don’t know much about the MultiQC modules, but the log file couldn’t be picked up after removing
--output-prefix
from Old_version_classify.log file. However, I was able to generate the multiqc report for Latest_version_vlassify.log using the code suggested by ChatGPT
j
So if you manually added output prefix it worked?
l
ganon.py
🤔 1
Yes i have tested manually by adding
--output-prefix       2612_pe_ERR5766176_db1.ganon
and it worked without modify
ganon.py
file
j
Ugh OK then I just don't understand MultiQC
I guess your suggestion makes sense then, if no output prefix then take the file name
Ooohhh
Did I generate it with
--verbose????
l
Great point. Indeed you used
--verbose
j
Does it work with the pipeline if we include that under flag
l
No, you still need to fix the missing pattern:
Copy code
removed with --min-count
j
Gah
l
Unfortunately we couldn’t include ganon to the multiqc report in this release. But I added nanoq to the report.
j
Ok no worries, that's fine