v1.4.1

.github/pull_request_template.md

@@ -38,3 +38,12 @@ Then,
 1. Open a PR from development to main with these changes.
 2. Wait for a review and merge.
 3. Create a new release on GitHub with the version number. Update the release notes with the changes made in this version.
+4. If the Docker github action fails (e.g., no space left on device), you can do it manually:
+   - Enable Docker buildx locally. If you don't have a builder that supports multi-arch, create one:
+   ```{bash}
+   docker buildx ls
+   docker buildx create --use --name multiarch-builder
+   docker buildx inspect --bootstrap
+   ```
+   - Login to ghcr `docker login --username <GitHub-Username> --password <personal-access-token> ghcr.io`
+   - Build the multi-platform image, insert the correct version `docker buildx build --platform linux/amd64,linux/arm64 -t ghcr.io/daisybio/drevalpy:v<INSERT-VERSION> -t ghcr.io/daisybio/drevalpy:latest --push .`

.github/workflows/build_package.yml

@@ -12,7 +12,7 @@ jobs:
         python: ["3.11", "3.12", "3.13"]
 
     steps:
-      - uses: actions/checkout@v5
+      - uses: actions/checkout@v6
         name: Check out source-code repository
 
       - name: Setup Python

.github/workflows/labeler.yml

@@ -10,7 +10,7 @@ jobs:
     runs-on: ubuntu-latest
     steps:
       - name: Check out the repository
-        uses: actions/checkout@v5
+        uses: actions/checkout@v6
 
       - name: Run Labeler
         uses: crazy-max/ghaction-github-labeler@v5.3.0

.github/workflows/publish-docker.yml

@@ -26,7 +26,7 @@ jobs:
     steps:
       # Necessary for buildx
       - name: Checkout repository
-        uses: actions/checkout@v5
+        uses: actions/checkout@v6
 
       - name: Setup QEMU
         uses: docker/setup-qemu-action@v3

.github/workflows/publish_docs.yml

@@ -7,7 +7,7 @@ jobs:
     runs-on: ubuntu-latest
 
     steps:
-      - uses: actions/checkout@v5
+      - uses: actions/checkout@v6
         name: Check out source-code repository
 
       - name: Setup Python

.github/workflows/python-package.yml

@@ -19,7 +19,7 @@ jobs:
 
     steps:
       - name: Check out the repository
-        uses: actions/checkout@v5
+        uses: actions/checkout@v6
 
       - name: Set up Python ${{ matrix.python-version }}
         uses: actions/setup-python@v6

.github/workflows/python-publish.yml

@@ -20,7 +20,7 @@ jobs:
     runs-on: ubuntu-latest
 
     steps:
-      - uses: actions/checkout@v5
+      - uses: actions/checkout@v6
       - name: Set up Python
         uses: actions/setup-python@v6
         with:

.github/workflows/run_tests.yml

@@ -28,7 +28,7 @@ jobs:
 
     steps:
       - name: Check out the repository
-        uses: actions/checkout@v5
+        uses: actions/checkout@v6
 
       - name: Set up Python ${{ matrix.python-version }}
         uses: actions/setup-python@v6
@@ -64,7 +64,7 @@ jobs:
           print("::set-output name=result::{}".format(result))
 
       - name: Restore pre-commit cache
-        uses: actions/cache@v4.3.0
+        uses: actions/cache@v5.0.1
         if: matrix.session == 'pre-commit'
         with:
           path: ~/.cache/pre-commit
@@ -78,15 +78,15 @@ jobs:
 
       - name: Upload coverage data
         if: always() && matrix.session == 'tests' && matrix.os == 'ubuntu-latest'
-        uses: actions/upload-artifact@v5
+        uses: actions/upload-artifact@v6
         with:
           name: coverage-data
           path: ".coverage.*"
           include-hidden-files: "true"
 
       - name: Upload documentation
         if: matrix.session == 'docs-build'
-        uses: actions/upload-artifact@v5
+        uses: actions/upload-artifact@v6
         with:
           name: docs
           path: docs/_build
@@ -96,7 +96,7 @@ jobs:
     needs: tests
     steps:
       - name: Check out the repository
-        uses: actions/checkout@v5
+        uses: actions/checkout@v6
 
       - name: Set up Python 3.13
         uses: actions/setup-python@v6
@@ -117,7 +117,7 @@ jobs:
           nox --version
 
       - name: Download coverage data
-        uses: actions/download-artifact@v5
+        uses: actions/download-artifact@v7
         with:
           name: coverage-data
 
@@ -128,6 +128,6 @@ jobs:
         run: nox --force-color --session=coverage -- xml -i
 
       - name: Upload coverage report
-        uses: codecov/codecov-action@v5.5.1
+        uses: codecov/codecov-action@v5.5.2
         with:
           token: ${{ secrets.CODECOV_TOKEN }}

.gitignore

@@ -10,6 +10,11 @@ data/TOYv2
 data/CTRPv1
 data/CTRPv2
 data/meta
+data/BeatAML2
+data/PDX_Bruna
+
+# Results directory is created when running the demo notebook
+results/
 
 # Byte-compiled / optimized / DLL files
 __pycache__/

.pre-commit-config.yaml

@@ -36,7 +36,7 @@ repos:
         types: [python]
         require_serial: true
         args:
-          - --ignore=D212,W503,C901,N803,N806,S615
+          - --ignore=D212,W503,C901,N803,N806,S615,S403,S301
       - id: pyupgrade
         name: pyupgrade
         description: Automatically upgrade syntax for newer versions.

README.md

@@ -37,6 +37,14 @@ Use DrEval to build drug response models that have an impact
 
 ---
 
+<picture>
+  <source media="(prefers-color-scheme: dark)" srcset="docs/_static/img/leaderboard_dark.png?v=4">
+  <source media="(prefers-color-scheme: light)" srcset="docs/_static/img/leaderboard_light.png?v=4">
+  <img alt="DrEvalPy Leaderboard" src="docs/_static/img/leaderboard_dark.png?v=4">
+</picture>
+
+---
+
 This project is a collaboration of the Technical University of Munich (TUM, Germany)
 and the Freie Universität Berlin (FU, Germany).
 

README.rst

@@ -68,3 +68,10 @@ Use DrEval to build drug response models that have an impact
 
 This project is a collaboration of the Technical University of Munich (TUM, Germany)
 and the Freie Universität Berlin (FU, Germany).
+
+Leaderboard
+-----------
+
+.. image:: docs/_static/img/leaderboard.png
+   :alt: DrEvalPy Leaderboard
+   :align: center

docs/_key_contributors.rst

@@ -5,4 +5,5 @@
    * `Mario Picciani <https://github.com/picciama>`_: Developer
    * `Markus List <https://github.com/mlist>`_: Advisor and PI of Data Science in Systems Biology, TUM
    * `Katharina Baum <https://github.com/kathbaum>`_: Advisor and PI of Data Integration in the Life Sciences, FU Berlin
-   * `Mathias Wilhelm <https://github.com/mwilhelm42>`_: Advisor and PI of Computational Mass Spectrometry, TUM
+   * `Mathias Wilhelm <https://github.com/mwilhelm42>`_: Advisor and PI of Computational Mass Spectrometry, TUM
+   * `Nico Trummer <https://github.com/nictru>`_: Contributor, `Orakl Oncology <https://www.orakl-oncology.com/>`_

docs/conf.py

@@ -56,9 +56,9 @@
 # the built documents.
 #
 # The short X.Y version.
-version = "1.4.0"
+version = "1.4.1"
 # The full version, including alpha/beta/rc tags.
-release = "1.4.0"
+release = "1.4.1"
 
 # The language for content autogenerated by Sphinx. Refer to documentation
 # for a list of supported languages.

docs/requirements.txt

@@ -1,4 +1,4 @@
 sphinx-autobuild==2025.8.25 ; python_version >= "3.11" and python_version <= "3.13"
 sphinx-autodoc-typehints==3.5.2 ; python_version >= "3.11" and python_version <= "3.13"
-sphinx-click==6.1.0 ; python_version >= "3.11" and python_version <= "3.13"
+sphinx-click==6.2.0 ; python_version >= "3.11" and python_version <= "3.13"
 sphinx-rtd-theme==3.0.2 ; python_version >= "3.11" and python_version <= "3.13"

docs/usage.rst

@@ -121,14 +121,18 @@ reproducible manner. We offer three settings via the ``--test_mode`` parameter:
 An underlying issue is that drugs have a rather unique IC50 range. That means that by just predicting the mean IC50
 that a drug has in the training set (aggregated over all cell lines), you can already achieve a seemingly good
 prediction (as evaluated by naive R^2 or correlation metrics). This is why we also offer the possibility to compare your model to a **NaivePredictor** that predicts
-the mean IC50 of all drugs in the training set. We also offer two less naive predictors:
-**NaiveCellLineMeanPredictor** and **NaiveDrugMeanPredictor**. The former predicts the mean IC50 of a cell line in
-the training set and the latter predicts the mean IC50 of a drug in the training set.
-Finally, the strongest naive baseline is the **NaiveMeanEffectPredictor**
-which combines the effects of cell lines and drugs.
-It is equivalent to the **NaiveCellLineMeanPredictor** and **NaiveDrugMeanPredictor** for the LDO and LPO settings, respectively,
+the mean IC50 of all drugs in the training set. We also offer several less naive predictors:
+**NaiveCellLineMeanPredictor**, **NaiveDrugMeanPredictor**, **NaiveTissueMeanPredictor**, and **NaiveTissueDrugMeanPredictor**.
+The **NaiveCellLineMeanPredictor** predicts the mean IC50 of a cell line in the training set,
+the **NaiveDrugMeanPredictor** predicts the mean IC50 of a drug in the training set,
+the **NaiveTissueMeanPredictor** predicts the mean IC50 of a tissue in the training set,
+and the **NaiveTissueDrugMeanPredictor** predicts the mean IC50 per tissue-drug combination (aggregated across all cell lines with that tissue-drug pair).
+The **NaiveMeanEffectPredictor** combines the effects of cell lines and drugs.
+It is equivalent to the **NaiveCellLineMeanPredictor** and **NaiveDrugMeanPredictor** for the LDO and LCO settings, respectively,
 as test cell line effects and drug effects are unknown in these settings.
 
+In LCO, **NaiveTissueDrugMeanPredictor** is the strongest baseline, while in all other settings, **NaiveMeanEffectPredictor** is the strongest.
+
 Available Models
 ------------------
 In addition to the Naive Predictors, we offer a variety of more advanced **baseline models** and
@@ -149,6 +153,8 @@ For ``--models``, you can also perform randomization and robustness tests. The `
 +---------------------------------+----------------------------+--------------------------------------+--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
 | NaiveTissueMeanPredictor        | Baseline Method            | Multi-Drug Model                     | Predicts the mean response of the tissue in the training set.                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                              |
 +---------------------------------+----------------------------+--------------------------------------+--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
+| NaiveTissueDrugMeanPredictor    | Baseline Method            | Multi-Drug Model                     | Predicts the mean response per tissue-drug combination in the training set (aggregated across all cell lines with that tissue-drug pair). Falls back to the overall dataset mean for unseen combinations.                                                                                                                                                                                                                                                                                                                                                                                   |
++---------------------------------+----------------------------+--------------------------------------+--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
 | NaiveMeanEffectsPredictor       | Baseline Method            | Multi-Drug Model                     | Predicts using ANOVA-like mean effect model of cell lines and drugs                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                        |
 +---------------------------------+----------------------------+--------------------------------------+--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
 | ElasticNet                      | Baseline Method            | Multi-Drug Model                     | Fits an `Sklearn Elastic Net <https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.ElasticNet.html>`_, `Lasso <https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.Lasso.html>`_, or `Ridge <https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.Ridge.html>`_ model on gene expression data and drug fingerprints (concatenated input matrix).                                                                                                                                                                                    |
@@ -268,11 +274,14 @@ the available datasets in the previous section.
 
 **Raw viability data**
 
-* DrEvalPy expects a csv-formatted file in the location ``<path_data>/<dataset>/<dataset_name>_raw.csv`` (corresponding to the ``--path_data`` and ``--dataset_name`` options),
-  which contains the raw viability data in long format with the columns ["dose", "response", "sample", "drug"] and an optional "replicate" column.
-  If replicates are provided, the procedure will fit one curve per sample / drug pair using all replicates.
+*   DrEvalPy expects a csv-formatted file in the location ``<path_data>/<dataset>/<dataset_name>_raw.csv`` (corresponding to the ``--path_data`` and ``--dataset_name`` options), which contains the raw viability data in long format with the columns ["dose", "response", "sample", "drug"] and an optional "replicate" column. If replicates are provided, the procedure will fit one curve per sample / drug pair using all replicates.
+* **All dosages have to be provided in µM!** Drevalpy will compute the following response measures:
+    * pEC50_curvecurator: computed internally by CurveCurator. Is computed as -log10(EC50_curvecurator[M]).
+    * EC50_curvecurator: given in µM
+    * IC50_curvecurator: given in µM
+    * LN_IC50_curvecurator: computed from IC50_curvecurator
+    * AUC_curvecurator
 * The option ``--curve_curator_cores`` must be set. ``--no_refitting`` must not be set.
-* Available measures are ["AUC", "pEC50", "EC50", "IC50"].
 * DrEvalPy provides all results of the fitting in the same folder including the fitted curves in a file folder ``<path_data>/<dataset>/<dataset_name>.csv``
 
 **Prefit viability data**

drevalpy/cli.py

@@ -7,7 +7,3 @@ def cli_main():
     """Command line interface entry point for the drug response evaluation pipeline."""
     args = get_parser().parse_args()
     main(args)
-
-
-if __name__ == "__main__":
-    cli_main()