New tool : multiclassprediction #8041
Conversation
There was a problem hiding this comment.
All requirements should have a version.
There was a problem hiding this comment.
This would be cool to have. Are you working on a conda recipe?
There was a problem hiding this comment.
Hi @bernt-matthias , yes I am working on a recipe as well, but its getting some errors and I don't know why its not working . Can you have a look there if you get some time.
bioconda/bioconda-recipes#65121
Thanks !
There was a problem hiding this comment.
paths need to be single quoted
There was a problem hiding this comment.
I would prefer if the actual source code would be published separately. Is there a citation?
There was a problem hiding this comment.
The method, parameters and outputs need detailed description. Without citation this is rather a blackbox.
paulzierep
left a comment
There was a problem hiding this comment.
Change to multi omics capability.
More help needed.
| <xml name="requirements"> | ||
| <requirements> | ||
| <requirement type="package" version="@TOOL_VERSION@">multi-omics-classification</requirement> | ||
| <requirement type="package" version="1.5.2">scikit-learn</requirement> |
There was a problem hiding this comment.
I think we can remove all this requirements, they are all in the multi-omics-classification recipe
| @@ -0,0 +1,7 @@ | |||
| scikit-learn | |||
There was a problem hiding this comment.
this file can be removed, this is handled by conda
| @@ -0,0 +1,10 @@ | |||
| categories: | |||
| - Machine Learning | |||
| description: A FAIR Machine Learning Framework for Benchmarking, Feature Selection, and Hyperparameter Optimization of Machine Learning Models for Multiomics Data in Galaxy | |||
There was a problem hiding this comment.
| description: A FAIR Machine Learning Framework for Benchmarking, Feature Selection, and Hyperparameter Optimization of Machine Learning Models for Multiomics Data in Galaxy | |
| description: A Machine Learning Framework for Benchmarking, Feature Selection, and Hyperparameter Optimization of Machine Learning Models for Multiomics Data |
| description: A FAIR Machine Learning Framework for Benchmarking, Feature Selection, and Hyperparameter Optimization of Machine Learning Models for Multiomics Data in Galaxy | ||
| homepage_url: https://github.com/usegalaxy-eu/MultiOmicsClassification | ||
| long_description: | | ||
| The MultiClassPrediction tool is designed for multi-omics data analysis within the Galaxy platform, enabling reproducible and standardized model evaluation across diverse biological datasets. |
There was a problem hiding this comment.
explain what it does, the galaxy part can be removed, we know its in galaxy :)
|
|
||
| <inputs> | ||
|
|
||
| <param name="X" type="data" format="tabular" label="X: Input features" help="Feature matrix (independent variables used for prediction)"/> |
There was a problem hiding this comment.
explain input format, first row is feature names ?
| </inputs> | ||
|
|
||
| <outputs> | ||
| <data name="output" format="csv" label="Multiclass AUC results"/> |
| <param name="sampling_strategy" value="No Sampling"/> | ||
| <param name="grid_search" value="false"/> | ||
|
|
||
| <output name="output"> |
There was a problem hiding this comment.
please add real test output, also for the plots and compare the files
| MultiClassPredict | ||
| ================== | ||
|
|
||
| A FAIR Machine Learning Framework for Benchmarking, Feature Selection, and Hyperparameter Optimization of Machine Learning Models for Multiomics Data in Galaxy |
There was a problem hiding this comment.
we do not need FAIR and Galaxy ref, we know its in Galaxy which is FAIR :)
| @@ -0,0 +1,178 @@ | |||
| <tool id="multi-omics-classification" name="multi-omics-classification" version="@TOOL_VERSION@+galaxy@VERSION_SUFFIX@" profile="@PROFILE@"> | |||
| <description>Multi Class prediction </description> | |||
There was a problem hiding this comment.
| <description>Multi Class prediction </description> | |
| <description>Multiomics Classification</description> |
|
|
||
| <inputs> | ||
|
|
||
| <param name="X" type="data" format="tabular" label="X: Input features" help="Feature matrix (independent variables used for prediction)"/> |
There was a problem hiding this comment.
Actually, we could make this a real multi-omics tool by making this a repeat, similar to
<repeat name="omics" min="0" title="Multiple omics layers?">
<param name="train_omics" type="data" optional="true" format="tabular" label="Training omics data"/>
<param name="test_omics" type="data" optional="true" format="tabular" label="Test omics data"/>
<param name="assay" type="text" optional="true" label="What type of assay is your input?" help="This would be used as output name.">
<sanitizer invalid_char=" ">
<valid initial="string.letters">
<add value="_"/>
</valid>
</sanitizer>
</param>
</repeat>
Then we can run the tool for each omics type using this as the label.
In the command line, one would need to merge X from all omics types once to create the combined input.
paulzierep
left a comment
There was a problem hiding this comment.
can you give the test-data better names, like prediction_results.csv (tabular would be better), also for the plots
FOR CONTRIBUTOR:
[ *] This PR adds a new tool or tool collection
There are two labels that allow to ignore specific (false positive) tool linter errors:
skip-version-check: Use it if only a subset of the tools has been updated in a suite.skip-url-check: Use it if github CI sees 403 errors, but the URLs work.