MaldiBatchKit 0.2.1__tar.gz → 0.2.2__tar.gz

{maldibatchkit-0.2.1 → maldibatchkit-0.2.2/MaldiBatchKit.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: MaldiBatchKit
-Version: 0.2.1
+Version: 0.2.2
 Summary: Batch-effect correction methods for MALDI-TOF spectra in clinical AMR prediction workflows
 Author-email: Ettore Rocchi <ettore.rocchi3@unibo.it>
 License-Expression: MIT
@@ -42,7 +42,7 @@ Requires-Dist: sphinx>=7.0.0; extra == "docs"
 Requires-Dist: pydata-sphinx-theme>=0.15.0; extra == "docs"
 Requires-Dist: sphinx-autodoc-typehints>=2.0.0; extra == "docs"
 Requires-Dist: sphinx-design>=0.6.0; extra == "docs"
-Requires-Dist: sphinx-click>=5.0.0; extra == "docs"
+Requires-Dist: sphinxcontrib-typer>=0.5.0; extra == "docs"
 Requires-Dist: nbsphinx>=0.9.0; extra == "docs"
 Requires-Dist: ipykernel>=6.0.0; extra == "docs"
 Requires-Dist: ipywidgets>=8.0.0; extra == "docs"
@@ -62,7 +62,10 @@ Dynamic: license-file
 [![License](https://img.shields.io/github/license/EttoreRocchi/MaldiBatchKit)](https://github.com/EttoreRocchi/MaldiBatchKit/blob/main/LICENSE)
 
 <p align="center">
-  <img src="docs/maldibatchkit_logo.png" alt="MaldiBatchKit" width="320"/>
+  <picture>
+    <source media="(prefers-color-scheme: dark)" srcset="https://raw.githubusercontent.com/EttoreRocchi/MaldiBatchKit/main/docs/maldibatchkit_logo_dark.png">
+    <img src="https://raw.githubusercontent.com/EttoreRocchi/MaldiBatchKit/main/docs/maldibatchkit_logo.png" alt="MaldiBatchKit" width="320"/>
+  </picture>
 </p>
 
 <p align="center">
@@ -70,7 +73,7 @@ Dynamic: license-file
 </p>
 
 <p align="center">
-  <img src="docs/batch_correction.svg" alt="Animated scatter plot: before correction MALDI-TOF samples cluster by acquisition site; after MaldiBatchKit the site effect is removed and samples cluster by species" width="720"/>
+  <img src="https://raw.githubusercontent.com/EttoreRocchi/MaldiBatchKit/main/docs/batch_correction.svg" alt="Animated scatter plot: before correction MALDI-TOF samples cluster by acquisition site; after MaldiBatchKit the site effect is removed and samples cluster by species" width="720"/>
 </p>
 
 <p align="center">
@@ -83,7 +86,7 @@ Dynamic: license-file
   <a href="#citation">Citation</a>
 </p>
 
-MaldiBatchKit is part of the **MaldiSuite** ecosystem and complements [MaldiAMRKit](https://github.com/EttoreRocchi/MaldiAMRKit): where MaldiAMRKit handles preprocessing, alignment and AMR-aware evaluation, MaldiBatchKit focuses on the *harmonization* step, removing the inter-batch / inter-site shifts that plague multi-centre MALDI-TOF studies.
+MaldiBatchKit corrects the inter-batch and inter-site shifts of multi-centre MALDI-TOF studies. It is part of the **MaldiSuite** ecosystem: it can be used on its own, or together with [MaldiAMRKit](https://github.com/EttoreRocchi/MaldiAMRKit) and [MaldiDeepKit](https://github.com/EttoreRocchi/MaldiDeepKit).
 
 ## Installation
 
@@ -124,8 +127,8 @@ Visit the **MaldiSuite** landing page at <https://ettorerocchi.github.io/MaldiSu
   - `SpeciesAwareComBat` - convenience preset for ComBat-Fortin with `species` as the protected biological covariate.
 - **Diagnostics**: kBET, LISI, silhouette-by-batch, per-batch peak drift, per-batch TIC coefficient of variation, per-batch spectrum count, plus a combined `diagnostic_report` DataFrame summary.
 - **Method selection**: `AutoCorrector` exposes `method` as a settable hyperparameter so `GridSearchCV` can sweep across corrector families and let the downstream classifier metric (AUROC) decide. Ships with a `NoOpCorrector` so "do nothing" can sit on the candidate list as an honest baseline.
-- **Diagnostic benchmark**: `BatchCorrectionBenchmark` runs a fixed set of metrics across multiple correctors with stratified bootstrap CIs and a tidy `(method, metric, value, ci_lo, ci_hi)` summary, ready for paper-figure comparisons.
-- **Visualization**: UMAP before/after, per-batch peak-shape overlays, before/after bar charts.
+- **Diagnostic benchmark**: `BatchCorrectionBenchmark` runs a fixed set of metrics across multiple correctors with intervals from stratified subsamples and a tidy `(method, metric, value, ci_lo, ci_hi)` summary, ready for paper-figure comparisons.
+- **Visualisation**: UMAP before/after, per-batch peak-shape overlays, before/after bar charts.
 - **Integration adapter**: `MaldiSetAdapter` turns a `maldiamrkit.MaldiSet` into a corrected `MaldiSet` in one call.
 - **CLI**: `maldibatchkit correct ...` and `maldibatchkit diagnose ...`.
 
@@ -176,6 +179,8 @@ corrected_ds.X      # harmonised feature matrix
 corrected_ds.y      # AMR labels, unchanged
 ```
 
+The `*_column` arguments name columns of the metadata behind `ds.meta`; a missing column raises `KeyError`. A corrector uses only what it accepts: `quality_column` has no effect with `SpeciesAwareComBat`.
+
 ### CLI
 
 The CLI is organised as `maldibatchkit correct <method>` + `maldibatchkit diagnose`. Every method has its own subcommand with only the flags it actually uses:
@@ -250,7 +255,7 @@ from sklearn.pipeline import Pipeline
 from maldibatchkit import AutoCorrector
 
 pipe = Pipeline([
-    ("correct", AutoCorrector(batch=batch, discrete_covariates=species)),
+    ("correct", AutoCorrector(batch=batch, discrete_covariates=species, quality=snr)),
     ("clf", LogisticRegression(max_iter=1000)),
 ])
 grid = GridSearchCV(
@@ -347,26 +352,24 @@ from maldibatchkit.diagnostics import (
 )
 ```
 
-All metrics take the same `(X, batch)` signature. `diagnostic_report` composes them into a tidy DataFrame suitable for `plot_diagnostic_summary`.
+Most metrics take `(X, batch)`; `per_batch_spectrum_count` takes only `batch`. `diagnostic_report(before, after, batch)` composes them into a tidy DataFrame suitable for `plot_diagnostic_summary`.
 
-## MaldiSuite Ecosystem
+## MaldiSuite ecosystem
 
-MaldiBatchKit is the harmonisation package of the **MaldiSuite** ecosystem:
+MaldiBatchKit is one of the three packages of the **MaldiSuite** ecosystem:
 
 - **[MaldiAMRKit](https://github.com/EttoreRocchi/MaldiAMRKit)** - data model (`MaldiSpectrum`, `MaldiSet`), preprocessing, alignment, peak detection, differential analysis, and AMR-aware evaluation.
 - **MaldiBatchKit** (this package) - batch-effect correction and harmonisation for multi-centre / multi-instrument MALDI-TOF spectra.
 - **[MaldiDeepKit](https://github.com/EttoreRocchi/MaldiDeepKit)** - sklearn-compatible deep learning classifiers (MLP, CNN, ResNet, Transformer).
 
-The three packages share the `MaldiSet` / `MaldiSpectrum` data model and are designed to compose in a single end-to-end pipeline. Install the full suite with `pip install maldisuite`. Landing page: [MaldiSuite](<https://ettorerocchi.github.io/MaldiSuite/>).
+The three packages share the `MaldiSet` data model. Each can be used on its own, or combined with the others. Install the full suite with `pip install maldisuite`. Landing page: [MaldiSuite](<https://ettorerocchi.github.io/MaldiSuite/>).
 
 ## Citation
 
-If you use MaldiBatchKit in academic work please cite:
-
-> _Citation will be available soon._
+MaldiBatchKit is part of MaldiSuite, which is described in an Application Note currently under review at *Bioinformatics*. The full citation will be added here once the article is published.
 
-along with the upstream references for whichever methods you apply (Johnson 2007, Fortin 2018, Chen 2022, Ritchie 2015, Korsunsky 2019).
+Please also cite the upstream references for whichever methods you apply (Johnson 2007, Fortin 2018, Chen 2022, Ritchie 2015, Korsunsky 2019).
 
 ## License
 
-MIT. See [LICENSE](LICENSE).
+MIT. See [LICENSE](https://github.com/EttoreRocchi/MaldiBatchKit/blob/main/LICENSE).

{maldibatchkit-0.2.1 → maldibatchkit-0.2.2}/MaldiBatchKit.egg-info/requires.txt

@@ -20,7 +20,7 @@ sphinx>=7.0.0
 pydata-sphinx-theme>=0.15.0
 sphinx-autodoc-typehints>=2.0.0
 sphinx-design>=0.6.0
-sphinx-click>=5.0.0
+sphinxcontrib-typer>=0.5.0
 nbsphinx>=0.9.0
 ipykernel>=6.0.0
 ipywidgets>=8.0.0

{maldibatchkit-0.2.1/MaldiBatchKit.egg-info → maldibatchkit-0.2.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: MaldiBatchKit
-Version: 0.2.1
+Version: 0.2.2
 Summary: Batch-effect correction methods for MALDI-TOF spectra in clinical AMR prediction workflows
 Author-email: Ettore Rocchi <ettore.rocchi3@unibo.it>
 License-Expression: MIT
@@ -42,7 +42,7 @@ Requires-Dist: sphinx>=7.0.0; extra == "docs"
 Requires-Dist: pydata-sphinx-theme>=0.15.0; extra == "docs"
 Requires-Dist: sphinx-autodoc-typehints>=2.0.0; extra == "docs"
 Requires-Dist: sphinx-design>=0.6.0; extra == "docs"
-Requires-Dist: sphinx-click>=5.0.0; extra == "docs"
+Requires-Dist: sphinxcontrib-typer>=0.5.0; extra == "docs"
 Requires-Dist: nbsphinx>=0.9.0; extra == "docs"
 Requires-Dist: ipykernel>=6.0.0; extra == "docs"
 Requires-Dist: ipywidgets>=8.0.0; extra == "docs"
@@ -62,7 +62,10 @@ Dynamic: license-file
 [![License](https://img.shields.io/github/license/EttoreRocchi/MaldiBatchKit)](https://github.com/EttoreRocchi/MaldiBatchKit/blob/main/LICENSE)
 
 <p align="center">
-  <img src="docs/maldibatchkit_logo.png" alt="MaldiBatchKit" width="320"/>
+  <picture>
+    <source media="(prefers-color-scheme: dark)" srcset="https://raw.githubusercontent.com/EttoreRocchi/MaldiBatchKit/main/docs/maldibatchkit_logo_dark.png">
+    <img src="https://raw.githubusercontent.com/EttoreRocchi/MaldiBatchKit/main/docs/maldibatchkit_logo.png" alt="MaldiBatchKit" width="320"/>
+  </picture>
 </p>
 
 <p align="center">
@@ -70,7 +73,7 @@ Dynamic: license-file
 </p>
 
 <p align="center">
-  <img src="docs/batch_correction.svg" alt="Animated scatter plot: before correction MALDI-TOF samples cluster by acquisition site; after MaldiBatchKit the site effect is removed and samples cluster by species" width="720"/>
+  <img src="https://raw.githubusercontent.com/EttoreRocchi/MaldiBatchKit/main/docs/batch_correction.svg" alt="Animated scatter plot: before correction MALDI-TOF samples cluster by acquisition site; after MaldiBatchKit the site effect is removed and samples cluster by species" width="720"/>
 </p>
 
 <p align="center">
@@ -83,7 +86,7 @@ Dynamic: license-file
   <a href="#citation">Citation</a>
 </p>
 
-MaldiBatchKit is part of the **MaldiSuite** ecosystem and complements [MaldiAMRKit](https://github.com/EttoreRocchi/MaldiAMRKit): where MaldiAMRKit handles preprocessing, alignment and AMR-aware evaluation, MaldiBatchKit focuses on the *harmonization* step, removing the inter-batch / inter-site shifts that plague multi-centre MALDI-TOF studies.
+MaldiBatchKit corrects the inter-batch and inter-site shifts of multi-centre MALDI-TOF studies. It is part of the **MaldiSuite** ecosystem: it can be used on its own, or together with [MaldiAMRKit](https://github.com/EttoreRocchi/MaldiAMRKit) and [MaldiDeepKit](https://github.com/EttoreRocchi/MaldiDeepKit).
 
 ## Installation
 
@@ -124,8 +127,8 @@ Visit the **MaldiSuite** landing page at <https://ettorerocchi.github.io/MaldiSu
   - `SpeciesAwareComBat` - convenience preset for ComBat-Fortin with `species` as the protected biological covariate.
 - **Diagnostics**: kBET, LISI, silhouette-by-batch, per-batch peak drift, per-batch TIC coefficient of variation, per-batch spectrum count, plus a combined `diagnostic_report` DataFrame summary.
 - **Method selection**: `AutoCorrector` exposes `method` as a settable hyperparameter so `GridSearchCV` can sweep across corrector families and let the downstream classifier metric (AUROC) decide. Ships with a `NoOpCorrector` so "do nothing" can sit on the candidate list as an honest baseline.
-- **Diagnostic benchmark**: `BatchCorrectionBenchmark` runs a fixed set of metrics across multiple correctors with stratified bootstrap CIs and a tidy `(method, metric, value, ci_lo, ci_hi)` summary, ready for paper-figure comparisons.
-- **Visualization**: UMAP before/after, per-batch peak-shape overlays, before/after bar charts.
+- **Diagnostic benchmark**: `BatchCorrectionBenchmark` runs a fixed set of metrics across multiple correctors with intervals from stratified subsamples and a tidy `(method, metric, value, ci_lo, ci_hi)` summary, ready for paper-figure comparisons.
+- **Visualisation**: UMAP before/after, per-batch peak-shape overlays, before/after bar charts.
 - **Integration adapter**: `MaldiSetAdapter` turns a `maldiamrkit.MaldiSet` into a corrected `MaldiSet` in one call.
 - **CLI**: `maldibatchkit correct ...` and `maldibatchkit diagnose ...`.
 
@@ -176,6 +179,8 @@ corrected_ds.X      # harmonised feature matrix
 corrected_ds.y      # AMR labels, unchanged
 ```
 
+The `*_column` arguments name columns of the metadata behind `ds.meta`; a missing column raises `KeyError`. A corrector uses only what it accepts: `quality_column` has no effect with `SpeciesAwareComBat`.
+
 ### CLI
 
 The CLI is organised as `maldibatchkit correct <method>` + `maldibatchkit diagnose`. Every method has its own subcommand with only the flags it actually uses:
@@ -250,7 +255,7 @@ from sklearn.pipeline import Pipeline
 from maldibatchkit import AutoCorrector
 
 pipe = Pipeline([
-    ("correct", AutoCorrector(batch=batch, discrete_covariates=species)),
+    ("correct", AutoCorrector(batch=batch, discrete_covariates=species, quality=snr)),
     ("clf", LogisticRegression(max_iter=1000)),
 ])
 grid = GridSearchCV(
@@ -347,26 +352,24 @@ from maldibatchkit.diagnostics import (
 )
 ```
 
-All metrics take the same `(X, batch)` signature. `diagnostic_report` composes them into a tidy DataFrame suitable for `plot_diagnostic_summary`.
+Most metrics take `(X, batch)`; `per_batch_spectrum_count` takes only `batch`. `diagnostic_report(before, after, batch)` composes them into a tidy DataFrame suitable for `plot_diagnostic_summary`.
 
-## MaldiSuite Ecosystem
+## MaldiSuite ecosystem
 
-MaldiBatchKit is the harmonisation package of the **MaldiSuite** ecosystem:
+MaldiBatchKit is one of the three packages of the **MaldiSuite** ecosystem:
 
 - **[MaldiAMRKit](https://github.com/EttoreRocchi/MaldiAMRKit)** - data model (`MaldiSpectrum`, `MaldiSet`), preprocessing, alignment, peak detection, differential analysis, and AMR-aware evaluation.
 - **MaldiBatchKit** (this package) - batch-effect correction and harmonisation for multi-centre / multi-instrument MALDI-TOF spectra.
 - **[MaldiDeepKit](https://github.com/EttoreRocchi/MaldiDeepKit)** - sklearn-compatible deep learning classifiers (MLP, CNN, ResNet, Transformer).
 
-The three packages share the `MaldiSet` / `MaldiSpectrum` data model and are designed to compose in a single end-to-end pipeline. Install the full suite with `pip install maldisuite`. Landing page: [MaldiSuite](<https://ettorerocchi.github.io/MaldiSuite/>).
+The three packages share the `MaldiSet` data model. Each can be used on its own, or combined with the others. Install the full suite with `pip install maldisuite`. Landing page: [MaldiSuite](<https://ettorerocchi.github.io/MaldiSuite/>).
 
 ## Citation
 
-If you use MaldiBatchKit in academic work please cite:
-
-> _Citation will be available soon._
+MaldiBatchKit is part of MaldiSuite, which is described in an Application Note currently under review at *Bioinformatics*. The full citation will be added here once the article is published.
 
-along with the upstream references for whichever methods you apply (Johnson 2007, Fortin 2018, Chen 2022, Ritchie 2015, Korsunsky 2019).
+Please also cite the upstream references for whichever methods you apply (Johnson 2007, Fortin 2018, Chen 2022, Ritchie 2015, Korsunsky 2019).
 
 ## License
 
-MIT. See [LICENSE](LICENSE).
+MIT. See [LICENSE](https://github.com/EttoreRocchi/MaldiBatchKit/blob/main/LICENSE).

{maldibatchkit-0.2.1 → maldibatchkit-0.2.2}/README.md

@@ -9,7 +9,10 @@
 [![License](https://img.shields.io/github/license/EttoreRocchi/MaldiBatchKit)](https://github.com/EttoreRocchi/MaldiBatchKit/blob/main/LICENSE)
 
 <p align="center">
-  <img src="docs/maldibatchkit_logo.png" alt="MaldiBatchKit" width="320"/>
+  <picture>
+    <source media="(prefers-color-scheme: dark)" srcset="https://raw.githubusercontent.com/EttoreRocchi/MaldiBatchKit/main/docs/maldibatchkit_logo_dark.png">
+    <img src="https://raw.githubusercontent.com/EttoreRocchi/MaldiBatchKit/main/docs/maldibatchkit_logo.png" alt="MaldiBatchKit" width="320"/>
+  </picture>
 </p>
 
 <p align="center">
@@ -17,7 +20,7 @@
 </p>
 
 <p align="center">
-  <img src="docs/batch_correction.svg" alt="Animated scatter plot: before correction MALDI-TOF samples cluster by acquisition site; after MaldiBatchKit the site effect is removed and samples cluster by species" width="720"/>
+  <img src="https://raw.githubusercontent.com/EttoreRocchi/MaldiBatchKit/main/docs/batch_correction.svg" alt="Animated scatter plot: before correction MALDI-TOF samples cluster by acquisition site; after MaldiBatchKit the site effect is removed and samples cluster by species" width="720"/>
 </p>
 
 <p align="center">
@@ -30,7 +33,7 @@
   <a href="#citation">Citation</a>
 </p>
 
-MaldiBatchKit is part of the **MaldiSuite** ecosystem and complements [MaldiAMRKit](https://github.com/EttoreRocchi/MaldiAMRKit): where MaldiAMRKit handles preprocessing, alignment and AMR-aware evaluation, MaldiBatchKit focuses on the *harmonization* step, removing the inter-batch / inter-site shifts that plague multi-centre MALDI-TOF studies.
+MaldiBatchKit corrects the inter-batch and inter-site shifts of multi-centre MALDI-TOF studies. It is part of the **MaldiSuite** ecosystem: it can be used on its own, or together with [MaldiAMRKit](https://github.com/EttoreRocchi/MaldiAMRKit) and [MaldiDeepKit](https://github.com/EttoreRocchi/MaldiDeepKit).
 
 ## Installation
 
@@ -71,8 +74,8 @@ Visit the **MaldiSuite** landing page at <https://ettorerocchi.github.io/MaldiSu
   - `SpeciesAwareComBat` - convenience preset for ComBat-Fortin with `species` as the protected biological covariate.
 - **Diagnostics**: kBET, LISI, silhouette-by-batch, per-batch peak drift, per-batch TIC coefficient of variation, per-batch spectrum count, plus a combined `diagnostic_report` DataFrame summary.
 - **Method selection**: `AutoCorrector` exposes `method` as a settable hyperparameter so `GridSearchCV` can sweep across corrector families and let the downstream classifier metric (AUROC) decide. Ships with a `NoOpCorrector` so "do nothing" can sit on the candidate list as an honest baseline.
-- **Diagnostic benchmark**: `BatchCorrectionBenchmark` runs a fixed set of metrics across multiple correctors with stratified bootstrap CIs and a tidy `(method, metric, value, ci_lo, ci_hi)` summary, ready for paper-figure comparisons.
-- **Visualization**: UMAP before/after, per-batch peak-shape overlays, before/after bar charts.
+- **Diagnostic benchmark**: `BatchCorrectionBenchmark` runs a fixed set of metrics across multiple correctors with intervals from stratified subsamples and a tidy `(method, metric, value, ci_lo, ci_hi)` summary, ready for paper-figure comparisons.
+- **Visualisation**: UMAP before/after, per-batch peak-shape overlays, before/after bar charts.
 - **Integration adapter**: `MaldiSetAdapter` turns a `maldiamrkit.MaldiSet` into a corrected `MaldiSet` in one call.
 - **CLI**: `maldibatchkit correct ...` and `maldibatchkit diagnose ...`.
 
@@ -123,6 +126,8 @@ corrected_ds.X      # harmonised feature matrix
 corrected_ds.y      # AMR labels, unchanged
 ```
 
+The `*_column` arguments name columns of the metadata behind `ds.meta`; a missing column raises `KeyError`. A corrector uses only what it accepts: `quality_column` has no effect with `SpeciesAwareComBat`.
+
 ### CLI
 
 The CLI is organised as `maldibatchkit correct <method>` + `maldibatchkit diagnose`. Every method has its own subcommand with only the flags it actually uses:
@@ -197,7 +202,7 @@ from sklearn.pipeline import Pipeline
 from maldibatchkit import AutoCorrector
 
 pipe = Pipeline([
-    ("correct", AutoCorrector(batch=batch, discrete_covariates=species)),
+    ("correct", AutoCorrector(batch=batch, discrete_covariates=species, quality=snr)),
     ("clf", LogisticRegression(max_iter=1000)),
 ])
 grid = GridSearchCV(
@@ -294,26 +299,24 @@ from maldibatchkit.diagnostics import (
 )
 ```
 
-All metrics take the same `(X, batch)` signature. `diagnostic_report` composes them into a tidy DataFrame suitable for `plot_diagnostic_summary`.
+Most metrics take `(X, batch)`; `per_batch_spectrum_count` takes only `batch`. `diagnostic_report(before, after, batch)` composes them into a tidy DataFrame suitable for `plot_diagnostic_summary`.
 
-## MaldiSuite Ecosystem
+## MaldiSuite ecosystem
 
-MaldiBatchKit is the harmonisation package of the **MaldiSuite** ecosystem:
+MaldiBatchKit is one of the three packages of the **MaldiSuite** ecosystem:
 
 - **[MaldiAMRKit](https://github.com/EttoreRocchi/MaldiAMRKit)** - data model (`MaldiSpectrum`, `MaldiSet`), preprocessing, alignment, peak detection, differential analysis, and AMR-aware evaluation.
 - **MaldiBatchKit** (this package) - batch-effect correction and harmonisation for multi-centre / multi-instrument MALDI-TOF spectra.
 - **[MaldiDeepKit](https://github.com/EttoreRocchi/MaldiDeepKit)** - sklearn-compatible deep learning classifiers (MLP, CNN, ResNet, Transformer).
 
-The three packages share the `MaldiSet` / `MaldiSpectrum` data model and are designed to compose in a single end-to-end pipeline. Install the full suite with `pip install maldisuite`. Landing page: [MaldiSuite](<https://ettorerocchi.github.io/MaldiSuite/>).
+The three packages share the `MaldiSet` data model. Each can be used on its own, or combined with the others. Install the full suite with `pip install maldisuite`. Landing page: [MaldiSuite](<https://ettorerocchi.github.io/MaldiSuite/>).
 
 ## Citation
 
-If you use MaldiBatchKit in academic work please cite:
-
-> _Citation will be available soon._
+MaldiBatchKit is part of MaldiSuite, which is described in an Application Note currently under review at *Bioinformatics*. The full citation will be added here once the article is published.
 
-along with the upstream references for whichever methods you apply (Johnson 2007, Fortin 2018, Chen 2022, Ritchie 2015, Korsunsky 2019).
+Please also cite the upstream references for whichever methods you apply (Johnson 2007, Fortin 2018, Chen 2022, Ritchie 2015, Korsunsky 2019).
 
 ## License
 
-MIT. See [LICENSE](LICENSE).
+MIT. See [LICENSE](https://github.com/EttoreRocchi/MaldiBatchKit/blob/main/LICENSE).

{maldibatchkit-0.2.1 → maldibatchkit-0.2.2}/maldibatchkit/__init__.py

@@ -58,7 +58,7 @@ from .metrics import (
     make_batch_scorer,
 )
 
-__version__ = "0.2.1"
+__version__ = "0.2.2"
 __author__ = "Ettore Rocchi"
 
 __all__ = [

{maldibatchkit-0.2.1 → maldibatchkit-0.2.2}/maldibatchkit/cli.py

@@ -357,6 +357,15 @@ def correct_limma(
     """limma::removeBatchEffect (Ritchie 2015)."""
     X, batch, extras = load_matrix(input, batch_csv)
     design = load_dataframe_csv(design_csv, X.index) if design_csv else None
+    if design is not None:
+        text = [
+            c for c in design.columns if not pd.api.types.is_numeric_dtype(design[c])
+        ]
+        if text:
+            raise typer.BadParameter(
+                f"--design-csv columns must be numeric; {text} are not. "
+                "Encode categorical covariates as dummy columns first."
+            )
     transformer = Limma(batch=batch, design=design, eps=eps)
     _run_and_save(transformer, X, output, batch, extras)
 

{maldibatchkit-0.2.1 → maldibatchkit-0.2.2}/maldibatchkit/corrections/baselines.py

@@ -43,9 +43,9 @@ class MedianCentering(BaseBatchCorrector):
     The fitted parameter is a ``(n_batches, n_features)`` table of
     medians. At ``transform`` time, the median for the batch each sample
     belongs to is subtracted. Batches that were absent at ``fit`` time
-    fall back to the grand median learned from training data (with a
-    warning); this keeps the transformer defined on unseen folds without
-    leaking information from the test set.
+    fall back to the grand median learned from training data; this keeps
+    the transformer defined on unseen folds without leaking information
+    from the test set.
 
     Parameters
     ----------

{maldibatchkit-0.2.1 → maldibatchkit-0.2.2}/maldibatchkit/corrections/maldi.py

@@ -22,7 +22,7 @@ def _try_import_maldiamrkit():
             "This transformer requires 'maldiamrkit' (normally installed "
             "alongside MaldiBatchKit). Reinstall with "
             "`pip install -U maldibatchkit` or "
-            "`pip install maldiamrkit>=0.13.0`."
+            "`pip install maldiamrkit>=0.15.0`."
         ) from exc
     return maldiamrkit
 

{maldibatchkit-0.2.1 → maldibatchkit-0.2.2}/maldibatchkit/metrics.py

@@ -54,11 +54,12 @@ and untouched. ``weights=`` only controls the *across-batch* reducer.
 
 Degenerate folds
 ----------------
-When a batch contains only one class in the held-out fold, AUROC and
-average precision are undefined; F1 / precision / recall fall back to
-zero. These batches are **silently dropped from the aggregate with a
-warning** and the remaining weights are renormalised. Use ``weights=``
-plus a careful split design if you need a different policy.
+A batch with fewer than two samples, or, for every metric except MCC,
+with only one class in the held-out fold, leaves the metric undefined.
+These batches are **dropped from the aggregate with a warning** and the
+remaining weights are renormalised (an array of weights keeps the
+entries of the remaining batches). Use ``weights=`` plus a careful split
+design if you need a different policy.
 
 ``make_batch_scorer`` factory
 -----------------------------
@@ -232,6 +233,15 @@ def _per_batch_metric(
             stacklevel=2,
         )
 
+    if not isinstance(weights, str | Mapping):
+        # An array is aligned with every batch level: keep the kept batches.
+        full = np.asarray(weights, dtype=float)
+        if full.shape != levels.shape:
+            raise ValueError(
+                f"weights array length {full.shape[0]} does not match the number "
+                f"of batch levels {levels.shape[0]}."
+            )
+        weights = full[valid]
     w = _resolve_weights(levels[valid], sizes[valid], weights)
     return float(np.sum(w * scores[valid]))
 

{maldibatchkit-0.2.1 → maldibatchkit-0.2.2}/requirements-docs.txt

@@ -5,7 +5,7 @@ sphinx>=7.0.0
 pydata-sphinx-theme>=0.15.0
 sphinx-autodoc-typehints>=2.0.0
 sphinx-design>=0.6.0
-sphinx-click>=5.0.0
+sphinxcontrib-typer>=0.5.0
 
 # Jupyter notebook support
 nbsphinx>=0.9.0

{maldibatchkit-0.2.1 → maldibatchkit-0.2.2}/tests/test_cli.py

@@ -256,6 +256,29 @@ def test_correct_limma(csv_inputs):
     assert result.exit_code == 0, result.output
 
 
+def test_correct_limma_text_design_refused(csv_inputs, tiny_dataset, tmp_path):
+    """A non-numeric design CSV gives a usage error, not a traceback."""
+    design = tmp_path / "text_design.csv"
+    _write_dataframe_csv(design, tiny_dataset["species"].rename("species").to_frame())
+    result = runner.invoke(
+        app,
+        [
+            "correct",
+            "limma",
+            "-i",
+            str(csv_inputs["X"]),
+            "-o",
+            str(csv_inputs["out"]),
+            "--batch-csv",
+            str(csv_inputs["batch"]),
+            "--design-csv",
+            str(design),
+        ],
+    )
+    assert result.exit_code == 2
+    assert "numeric" in " ".join(result.output.split())
+
+
 def test_correct_harmony(csv_inputs):
     result = runner.invoke(
         app,

{maldibatchkit-0.2.1 → maldibatchkit-0.2.2}/tests/test_metrics.py

@@ -376,3 +376,25 @@ def test_single_class_batch_dropped_for_f1_precision_recall(fn):
         with pytest.warns(UserWarning, match="dropped"):
             v = fn(y, y.copy(), batch=batch)
     assert v == pytest.approx(1.0)
+
+
+def test_array_weights_follow_dropped_batches():
+    """An array of weights stays aligned when a single-class batch is dropped."""
+    batch = np.array(["a"] * 10 + ["b"] * 10 + ["c"] * 10)
+    y = np.array([0] * 10 + [0, 1] * 5 + [0, 1] * 5)
+    score = np.linspace(0.0, 1.0, 30)
+    with pytest.warns(UserWarning, match="dropped"):
+        got = batch_roc_auc_score(y, score, batch=batch, weights=[5.0, 1.0, 3.0])
+    with pytest.warns(UserWarning, match="dropped"):
+        ref = batch_roc_auc_score(
+            y, score, batch=batch, weights={"a": 5.0, "b": 1.0, "c": 3.0}
+        )
+    assert got == pytest.approx(ref)
+
+
+def test_array_weights_wrong_length_raises():
+    batch = np.array(["a"] * 4 + ["b"] * 4 + ["c"] * 4)
+    y = np.array([0, 1] * 6)
+    score = np.linspace(0.0, 1.0, 12)
+    with pytest.raises(ValueError, match="does not match"):
+        batch_roc_auc_score(y, score, batch=batch, weights=[1.0, 1.0])