release v0.1.0

Author: quentinblampey

CHANGELOG.md

@@ -1,3 +1,3 @@
-## [0.1.0] - tbd
+## [0.1.0] - 2024-09-11
 
-tbd
+First official `novae` release. Preprint coming soon.

data/README.md

@@ -1,8 +1,15 @@
 # Public datasets
 
-We detail below how to download public spatial transcriptomics datasets. The data will be saved in this directory, and will be used to train `novae`.
+We detail below how to download public spatial transcriptomics datasets.
 
-## Download
+## Option 1: Hugging Face Hub
+
+We store our dataset on [Hugging Face Hub](https://huggingface.co/datasets/MICS-Lab/novae).
+To automatically download these slides, you can use the [`novae.utils.load_dataset`](https://mics-lab.github.io/novae/api/novae.utils/#novae.utils.load_dataset) function.
+
+NB: not all slides are uploaded on Hugging Face yet, but we are progressively adding new slides. To get the full dataset right now, use the "Option 2" below.
+
+## Option 2: Download
 
 For consistency, all the scripts below need to be executed at the root of the `data` directory (i.e., `novae/data`).
 
@@ -50,15 +57,15 @@ All above datasets can be downloaded using a single command line. Make sure you
 sh _scripts/1_download_all.sh
 ```
 
-## Preprocess and prepare for training
+### Preprocess and prepare for training
 
 The script bellow will copy all `adata.h5ad` files into a single directory, compute UMAPs, and minor preprocessing. See the `argparse` helper of this script for more details.
 
 ```sh
 python _scripts/2_prepare.py
 ```
 
-## Usage
+### Usage
 
 These datasets can be used during training (see the `scripts` directory at the root of the `novae` repository).
 

docs/api/novae.plot.md

@@ -5,3 +5,5 @@
 ::: novae.plot.pathway_scores
 
 ::: novae.plot.paga
+
+::: novae.plot.spatially_variable_genes

docs/tutorials/main_usage.ipynb

@@ -10,7 +10,7 @@
 from ._utils import get_categorical_color_palette
 
 
-def domains_proportions(adata: AnnData | list[AnnData], obs_key: str | None, figsize: tuple[int, int] = (2, 5)):
+def domains_proportions(adata: AnnData | list[AnnData], obs_key: str | None = None, figsize: tuple[int, int] = (2, 5)):
     """Show the proportion of each domain in the slide(s).
 
     Args:

novae/plot/_bar.py

@@ -82,6 +82,9 @@ def _domains_hierarchy(
 def paga(adata: AnnData, obs_key: str | None = None, **paga_plot_kwargs: int):
     """Plot a PAGA graph.
 
+    Info:
+        Currently, this function only supports one slide per call.
+
     Args:
         adata: An AnnData object.
         obs_key: Name of the key from `adata.obs` containing the Novae domains. By default, the last available domain key is shown.

novae/plot/_graph.py

@@ -71,6 +71,9 @@ def pathway_scores(
 ) -> pd.DataFrame | None:
     """Show a heatmap of pathway scores for each domain.
 
+    Info:
+        Currently, this function only supports one slide per call.
+
     Args:
         adata: An `AnnData` object.
         pathways: Either a dictionary of pathways (keys are pathway names, values are lists of gane names), or a path to a [GSEA](https://www.gsea-msigdb.org/gsea/msigdb/index.jsp) JSON file.

novae/plot/_heatmap.py

@@ -34,7 +34,6 @@ def domains(
     Info:
         Make sure you have already your Novae domains assigned to the `AnnData` object. You can use `model.assign_domains(...)` to do so.
 
-
     Args:
         adata: An `AnnData` object, or a list of `AnnData` objects.
         obs_key: Name of the key from `adata.obs` containing the Novae domains. By default, the last available domain key is shown.
@@ -113,7 +112,25 @@ def spatially_variable_genes(
     min_positive_ratio: float = 0.05,
     return_list: bool = False,
     **kwargs: int,
-) -> list[str]:
+) -> None | list[str]:
+    """Plot the most spatially variable genes (SVG) for a given `AnnData` object.
+
+    !!! info
+        Currently, this function only supports one slide per call.
+
+    Args:
+        adata: An `AnnData` object corresponding to one slide.
+        obs_key: Key in `adata.obs` that contains the domains. By default, it will use the last available Novae domain key.
+        top_k: Number of SVG to be shown.
+        show: Whether to show the plot.
+        cell_size: Size of the cells or spots (`spot_size` argument of `sc.pl.spatial`).
+        min_positive_ratio: Genes whose "ratio of cells expressing it" is lower than this threshold are not considered.
+        return_list: Whether to return the list of SVG instead of plotting them.
+        **kwargs: Additional arguments for `sc.pl.spatial`.
+
+    Returns:
+        A list of SVG names if `return_list` is `True`.
+    """
     assert isinstance(adata, AnnData), f"Received adata of type {type(adata)}. Currently only AnnData is supported."
 
     obs_key = utils.check_available_domains_key([adata], obs_key)

novae/plot/_spatial.py

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "novae"
-version = "0.0.5"
+version = "0.1.0"
 description = "Graph-based foundation model for spatial transcriptomics data"
 documentation = "https://mics-lab.github.io/novae/"
 homepage = "https://mics-lab.github.io/novae/"