Hi all, As part of my bachelor's thesis, I’ve dev...
# modules
j
Hi all, As part of my bachelor's thesis, I’ve developed a new module for the Detaxizer pipeline — it creates an HTML report page to visualize detected taxa. I’d really appreciate it if I could discuss it with someone and get feedback on whether it could be published or integrated into the nf-core ecosystem. Thanks in advance!
u
Hi @Jiří Vlasák I am keen to have a look. Is this module already in the nf-core/modules repo?
Do you have the code in a Github repo?
j
I got in my repo, I have not push it somewhere else yet. I will drop you a link
r
You could also reach out in the #detaxizer channel, if you haven't already
m
I only had a brief look at your repository, but to me it seems that your module is essentially ready to become a nf-core module. It already has a conda environment and uses suitable containers and therefore the most important technical requirements are fulfilled, thus you could start the process of opening a pull request to the modules repository. Modules like this usually have
custom
as part of their name, so searching for this on the modules page will yield some that may serve as template. Generally speaking, most nf-core pipelines use MultiQC for report generation, which is perfectly capable of rendering tables in various forms and also displaying custom graphics. Therefore, I suggest you pivot the focus of your module purely to generating a customizable figure of a phylogenetic tree instead of a whole HTML report. By exposing some of the styling as an input channel pipeline authors can devise a parameter to allow for easy customization when using your module in their pipelines. Lastly, I strongly recommend joining the #C080325T3MG channel, because this is where many people are discussing means to create and style pipeline reports.
j
Thank you for your advise. I will go through that this weekend.
Hello, Im sorry for longer pause, but I finally defended my thesis and completed my Final State Exams, so now im Bcs. I managed to upgraded my module. Now it creates Phylogenetic tree based on Kraken2 output (Here I ask for advice if i should switch to Bracken output or leave it as module dependent on Kraken2). I will start the process of openin a pull request to the modules ropository.
Here is a image of the phylogenetic tree, data were "given" to me and they were collected form cancer tissues.
image.png
Axis X is number of classification ranks starting at 0 (root) to end node.
Y is number of end nodes( meaning classified taxos with no more childs)
m
Firstly: Congratulations to your successful defense and graduation! party dino Regarding the Kraken/Bracken output, i suggest asking the #meta-omics people, who would also be the ones using your module eventually in their pipelines.