### Environment Setup YAML Configuration

Source: https://github.com/theislab/single-cell-best-practices/blob/main/scripts/dropdowns/env_setup.md

Example of a YAML configuration file for environment setup.

```yaml
dependencies:
  - python=3.9
  - pip
  - virtualenv

packages:
  - numpy
  - pandas
  - scikit-learn

settings:
  data_path: "/data"
  model_path: "/models"
  log_level: "INFO"
```

--------------------------------

### Setup and Track Lamindb Instance

Source: https://github.com/theislab/single-cell-best-practices/blob/main/scripts/conditions/differential_gene_expression_dataset.ipynb

Initializes lamindb settings and starts tracking notebook execution. Ensure the instance slug matches your project.

```python
import lamindb as ln

assert ln.setup.settings.instance.slug == "theislab/sc-best-practices"
ln.track()
```

--------------------------------

### Install pre-commit

Source: https://github.com/theislab/single-cell-best-practices/blob/main/CONTRIBUTING.md

Install the pre-commit tool for automatically checking code and markdown before commits.

```bash
pip install pre-commit
```

--------------------------------

### Default Environment Setup Text

Source: https://github.com/theislab/single-cell-best-practices/blob/main/scripts/dropdowns/env_setup.md

Provides the default text content for environment setup instructions.

```markdown
This section outlines the steps to set up your environment. It includes installing necessary dependencies, configuring your project, and preparing for development.

**1. Install Dependencies:**

```bash
pip install -r requirements.txt
```

**2. Configure Project:**

Ensure your configuration files are correctly set up. Refer to the `config.yaml` for details.

**3. Prepare for Development:**

Run the setup script to finalize the environment preparation.

```bash
python setup.py
```

**Note:** Always ensure you are using the correct Python version as specified in the project documentation.
```

--------------------------------

### View AnnData Setup for scvi-tools

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/cellular_structure/integration.ipynb

Displays the AnnData setup information for a scvi-tools model. This confirms the version of scvi-tools used and the setup status.

```python
model_scanvi.view_anndata_setup()
```

--------------------------------

### Install and Load FigR Package

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/chromatin_accessibility/gene_regulatory_networks_atac.ipynb

Installs the FigR package and its dependency BuenColors from GitHub if not already installed. This is a prerequisite for running the notebook.

```r
# the installation of this package is required for the proper execution of this notebook.
if(!suppressMessages(require("FigR"))){
    suppressMessages(devtools::install_github("caleblareau/BuenColors")) # the package BuenColors is also a devtools dependency to install FigR.
    suppressMessages(devtools::install_github("buenrostrolab/FigR"))
}
```

--------------------------------

### View AnnData Setup for scVI

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/cellular_structure/integration.ipynb

Display a detailed description of the AnnData object's setup for scVI, including parameters and keys used.

```python
model_scvi.view_anndata_setup()
```

--------------------------------

### Install cisTopic Package

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/chromatin_accessibility/gene_regulatory_networks_atac.ipynb

Installs the cisTopic package from GitHub using devtools. A kernel restart might be required after installation.

```r
if(!suppressMessages(require("cisTopic")))
    suppressMessages(devtools::install_github("aertslab/cisTopic"))
```

--------------------------------

### Display Setup Arguments for SCANVI

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/cellular_structure/integration.ipynb

Shows the arguments used during the `SCANVI.setup_anndata` process. This is helpful for debugging and verifying the AnnData preparation steps.

```python
print(model_scanvi.summary())
```

--------------------------------

### Python Environment Setup

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/introduction/advanced_data_structures_and_frameworks.ipynb

Sets up the Python environment by filtering warnings. This is useful for cleaning up output during data analysis.

```python
import os
import warnings

warnings.filterwarnings("ignore")
os.environ["PYTHONWARNINGS"] = "ignore"
```

--------------------------------

### Install SpatialData using pip

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/introduction/advanced_data_structures_and_frameworks.ipynb

Installs the SpatialData library using pip, a package installer for Python. This is one of the methods to get SpatialData ready for use.

```bash
pip install spatialdata
```

--------------------------------

### Example Project Folder Layout

Source: https://github.com/theislab/single-cell-best-practices/blob/main/CONTRIBUTING.md

Illustrates the directory structure of the single-cell best practices book, showing how chapters and their associated files are organized within section folders.

```bash
├── conditions
│	├── compositional_keytakeaways.txt
│	├── compositional.bib
│	├── compositional.ipynb
│	├── compositional.yml
│	├── differential_gene_expression_keytakeaways.txt
│	├── differential_gene_expression.bib
│	├── differential_gene_expression.ipynb
│	├── differential_gene_expression.yml
│	├── gsea_pathway_keytakeaways.txt
│	└── ...
├── ...
├── _toc.yml
├── _config.yml
├── acknowledgements.md
├── CHANGELOG.md
├── glossary.md
├── outlook.md
├── preamble.md
├── _static
│	├── book.css
│	├── book.js
│	├── favicon.ico
│	├── images
│	│	├── conditions
│	│	│	├── compositional.jpg
│	│	│	└── differential_gene_expression.jpg
│	│	└── ...

```

--------------------------------

### Example Chapter Folder Structure

Source: https://github.com/theislab/single-cell-best-practices/blob/main/CONTRIBUTING.md

Shows the essential files required for each chapter, including the notebook, bibliography, Conda environment file, and key takeaways summary.

```bash
├── SECTION-NAME
│	├── CHAPTER-NAME.ipynb
│	├── CHAPTER-NAME.bib
│	├── CHAPTER-NAME.yml
│	├── CHAPTER-NAME_keytakeaways.txt
│	├── ...

```

--------------------------------

### Install pre-commit in repository

Source: https://github.com/theislab/single-cell-best-practices/blob/main/CONTRIBUTING.md

Activate pre-commit hooks in the root of the repository to enforce code quality standards.

```bash
pre-commit install
```

--------------------------------

### Setup and Tracking Imports

Source: https://github.com/theislab/single-cell-best-practices/blob/main/scripts/trajectory/rna_velocity_dataset.ipynb

Imports necessary libraries and asserts the laminDB instance. Tracks the current run.

```python
from pathlib import Path

import lamindb as ln
import scvelo as scv

assert ln.setup.settings.instance.slug == "theislab/sc-best-practices"

ln.track("aWcFeMqzSw0X")
```

--------------------------------

### Install AnnData using pip

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/introduction/fundamental_data_structures_and_frameworks.ipynb

Install the AnnData library using pip. This is a common method for Python package management.

```bash
pip install anndata
```

--------------------------------

### Install lamindb Package

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/_static/default_text_lamindb_setup.md

Install the lamindb Python package using pip. This is the first step to using lamindb.

```bash
pip install lamindb
```

--------------------------------

### Import Libraries and Setup Scanpy

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/spatial/domains.ipynb

Imports necessary libraries (scanpy and squidpy) and sets up global configurations for scanpy, including verbosity and figure parameters. This is a standard setup for most spatial omics analyses using these libraries.

```python
import scanpy as sc
import squidpy as sq

sc.settings.verbosity = 3
sc.settings.set_figure_params(dpi=80, facecolor="white")
```

--------------------------------

### Create Working Directory and Download Data

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/introduction/raw_data_processing.md

Sets up a working directory, downloads an example dataset and a CB permit list, and decompresses them. The `&&` operator chains commands, and the `|` operator pipes output between commands.

```bash
# Create a working dir and go to the working directory
## The && operator helps execute two commands using a single line of code.
mkdir af_xmpl_run && cd af_xmpl_run

# Fetch the example dataset and CB permit list and decompress them
## The pipe operator (|) passes the output of the wget command to the tar command.
## The dash operator (-) after `tar xzf` captures the output of the first command.
## - example dataset
wget -qO- https://umd.box.com/shared/static/lx2xownlrhz3us8496tyu9c4dgade814.gz | tar xzf - --strip-components=1 -C .
## The fetched folder containing the fastq files is called toy_read_fastq.
fastq_dir="toy_read_fastq"
## The fetched folder containing the human ref files is called toy_human_ref.
ref_dir="toy_human_ref"

# Fetch CB permit list
## the right chevron (>) redirects the STDOUT to a file.
wget -qO- https://github.com/f0t1h/3M-february-2018/raw/master/3M-february-2018.txt.gz | gunzip - > 3M-february-2018.txt

```

--------------------------------

### Install lamindb with extensions

Source: https://github.com/theislab/single-cell-best-practices/blob/main/CONTRIBUTING.md

Install the lamindb Python package with bionty, jupyter, and zarr extensions for data management and analysis.

```bash
pip install lamindb[bionty,jupyter,zarr]
```

--------------------------------

### Create and Activate Conda Environment

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/introduction/raw_data_processing.md

Installs the Simpleaf package and its dependencies into a new conda environment named 'af'.

```bash
conda create -n af -y -c bioconda simpleaf
conda activate af
```

--------------------------------

### Install Scanpy using pip

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/introduction/fundamental_data_structures_and_frameworks.ipynb

Use this command to install the scanpy package via pip.

```bash
pip install scanpy
```

--------------------------------

### Setup AnnData for Cell2location

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/spatial/deconvolution.ipynb

Prepares the AnnData object for the Cell2location model by specifying the batch key. This is a prerequisite for model initialization.

```python
c2l.models.Cell2location.setup_anndata(
    adata=adata_st,
    batch_key="patient",
)
```

--------------------------------

### Install and Load BSgenome Package

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/chromatin_accessibility/gene_regulatory_networks_atac.ipynb

Ensures the BSgenome.Hsapiens.UCSC.hg38 package is installed and loaded for human genome annotations. Requires BiocManager for installation.

```r
if(!suppressMessages(require('BSgenome.Hsapiens.UCSC.hg38'))){
    if (!require("BiocManager", quietly = TRUE))
        install.packages("BiocManager")
    BiocManager::install("BSgenome.Hsapiens.UCSC.hg38")
}
```

```r
suppressMessages(library(BSgenome.Hsapiens.UCSC.hg38))
```

--------------------------------

### Connect to Lamindb Instance and Load Artifact

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/template/template.ipynb

Connects to a specified lamindb instance and loads a dataset artifact. Ensure lamindb is installed and configured.

```python
import lamindb as ln

assert ln.setup.settings.instance.slug == "theislab/sc-best-practices"

ln.track()
af = ln.Artifact.connect("theislab/sc-best-practices").get(
    key="introduction/interoperability_mdata.h5mu", is_latest=True
)
adata = af.load()
```

--------------------------------

### Install rapids-singlecell with CUDA 12

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/introduction/rapids_singlecell.ipynb

Install the rapids-singlecell package with CUDA 12 support and the full RAPIDS stack. The `--extra-index-url` is necessary as RAPIDS wheels are hosted on the NVIDIA PyPI index.

```bash
pip install 'rapids-singlecell-cu12[rapids]' --extra-index-url=https://pypi.nvidia.com
```

--------------------------------

### Install Muon

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/introduction/advanced_data_structures_and_frameworks.ipynb

Install the Muon package using pip. This command is used to add Muon to your Python environment for multimodal data analysis.

```bash
pip install muon
```

--------------------------------

### Setup R dependencies

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/conditions/gsea_pathway.ipynb

Activates R integration with Python using rpy2 and loads the SingleCellExperiment package in R.

```python
# Setting up R dependencies
import anndata2ri

%load_ext rpy2.ipython

anndata2ri.activate()

```

```R
suppressPackageStartupMessages({
    library(SingleCellExperiment)
})
```

--------------------------------

### Install AnnData using Conda

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/introduction/fundamental_data_structures_and_frameworks.ipynb

Install the AnnData library using Conda. This is useful for managing environments and dependencies, especially in scientific computing.

```bash
conda install -c conda-forge anndata
```

--------------------------------

### Install rapids-singlecell with CUDA 13

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/introduction/rapids_singlecell.ipynb

Install the rapids-singlecell package with CUDA 13 support and the full RAPIDS stack. The `--extra-index-url` is necessary as RAPIDS wheels are hosted on the NVIDIA PyPI index.

```bash
pip install 'rapids-singlecell-cu13[rapids]' --extra-index-url=https://pypi.nvidia.com
```

--------------------------------

### Setup AnnData for scVI

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/cellular_structure/integration.ipynb

Prepare the AnnData object for scVI by specifying the counts layer and batch key. This stores necessary information for the scVI model.

```python
scvi.model.SCVI.setup_anndata(adata_scvi, layer="counts", batch_key=batch_key)
```

--------------------------------

### Initialize lamindb and Track Run

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/cellular_structure/clustering.ipynb

Initializes the lamindb instance and starts tracking the current analysis run. This is essential for reproducibility and logging.

```python
import lamindb as ln
import scanpy as sc

assert ln.setup.settings.instance.slug == "theislab/sc-best-practices"

ln.track("rJhR7SskiROg")
```

--------------------------------

### Importing Libraries and Setting Up Warnings

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/air_repertoire/clonotype.ipynb

Imports necessary libraries for data analysis and sets up warning filters to suppress non-critical messages during execution. This is a common setup for data science workflows.

```python
import warnings

warnings.filterwarnings(
    "ignore",
    ".*IProgress not found*",
)
warnings.simplefilter(action="ignore", category=FutureWarning)

import numpy as np
import pandas as pd
import scanpy as sc
import scirpy as ir
from palmotif import compute_motif, svg_logo

warnings.simplefilter(action="ignore", category=pd.errors.DtypeWarning)
```

--------------------------------

### Setup AnnData for scGen

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/conditions/perturbation_modeling.ipynb

Configures the AnnData object for scGen, specifying the batch key and labels key. This is a mandatory step before using scGen.

```python
pt.tl.SCGEN.setup_anndata(adata_t, batch_key="condition", labels_key="cell_type")
```

--------------------------------

### Import Libraries and Track Run

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/conditions/differential_gene_expression.ipynb

Imports essential libraries for data analysis and tracking the notebook run with lamindb. Ensure these libraries are installed in your environment.

```python
import decoupler as dc
import lamindb as ln
import numpy as np
import pandas as pd
import pertpy as pt
import scanpy as sc

ln.track()
```

--------------------------------

### Setup AnnData for totalVI

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/multimodal_integration/paired_integration.ipynb

Configures the AnnData object for totalVI, specifying the protein expression key, RNA counts layer, and batch key for batch effect correction.

```python
scvi.model.TOTALVI.setup_anndata(
    adata,
    protein_expression_obsm_key="protein_expression",
    layer="counts",
    batch_key="batch",
)
```

--------------------------------

### Get lamindb Version

Source: https://github.com/theislab/single-cell-best-practices/blob/main/scripts/cellular_structure/integration_dataset.ipynb

Retrieves the currently installed version of the lamindb library.

```python
ln.__version__
```

--------------------------------

### Setup AnnData for totalVI Model

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/multimodal_integration/advanced_integration.ipynb

Configure the AnnData object for the totalVI model, specifying the layer for counts, the batch key, and the key for protein expression in the observations' metadata.

```python
TOTALVI.setup_anndata(
    adata,
    layer="counts",
    batch_key="batch",
    protein_expression_obsm_key="protein_counts",
)
```

--------------------------------

### Getting Image Channel Names

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/introduction/advanced_data_structures_and_frameworks.ipynb

This function returns the names or identifiers for each channel within an image. For example, it can indicate which staining or wavelength each channel corresponds to.

```python
sd.models.get_channel_names(sdata["raw_image"])
```

--------------------------------

### Import Libraries and Setup Scanpy

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/preprocessing_visualization/dimensionality_reduction.ipynb

Imports necessary Python libraries (lamindb, scanpy) and configures Scanpy settings for plotting and verbosity. It also tracks the current run with lamindb.

```python
import lamindb as ln
import scanpy as sc

# Suppress verbose logging from Scanpy
sc.settings.verbosity = 0

# Set figure parameters for clean, minimal plots
sc.settings.set_figure_params(dpi=80, facecolor="white", frameon=False)

assert ln.setup.settings.instance.slug == "theislab/sc-best-practices"

ln.track()
```

--------------------------------

### Setup AnnData for cell2location Regression Model

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/spatial/deconvolution.ipynb

Prepares the AnnData object for training the cell2location regression model by specifying batch keys, labels, covariates, and the layer containing counts.

```python
c2l.models.RegressionModel.setup_anndata(
    adata=adata_sc,
    batch_key="donor_id",
    labels_key="cell_type_original",
    categorical_covariate_keys=["assay"],
    layer="counts",
)
```

--------------------------------

### Install SpatialData using Conda

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/introduction/advanced_data_structures_and_frameworks.ipynb

Installs the SpatialData library using Conda, a cross-platform package and environment manager. This is an alternative installation method.

```bash
conda install -c conda-forge spatialdata
```

--------------------------------

### Initialize Environment and Load Data

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/cellular_structure/annotation.ipynb

Sets up the environment by connecting to a lamindb instance and tracking a notebook run. Loads a preprocessed AnnData object.

```python
import shutil
import sys
from pathlib import Path

import celltypist
import lamindb as ln
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd
import pandas.core.indexes.base as pandas_indexes_base
import scanpy as sc
import scarches as sca
import seaborn as sns
from celltypist import models
from scipy.sparse import csr_matrix

ln.connect("theislab/sc-best-practices")

ln.track("BnpvfJLWjHuE")
```

```python
sc.set_figure_params(figsize=(5, 5))
```

```python
af = ln.Artifact.get(key="cellular_structure/s4d8_clustered.h5ad", is_latest=True)
adata = af.load(is_run_input=False)
```

--------------------------------

### Initialize LaminDB and Load Data

Source: https://github.com/theislab/single-cell-best-practices/blob/main/scripts/dropdowns/lamin_setup.md

Sets up the LaminDB instance and loads data from a specified path. Ensure the path points to your data directory.

```python
import lamindb as la

# Initialize LaminDB
la.setup.init(
    "your_project_name",
    "your_instance_name",
    storage="/path/to/your/storage",
    db="postgresql://user:password@host:port/database",
)

# Load data from a directory
la.from_dir("/path/to/your/data")
```

--------------------------------

### Import Libraries and Setup

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/surface_protein/batch_correction.ipynb

Imports necessary libraries like warnings, muon, scanpy, and lamindb. It also sets global options for these libraries, such as disabling warnings, configuring Muon, and setting Scanpy's figure parameters. Finally, it initializes lamindb tracking.

```python
import warnings

import muon as mu
import scanpy as sc

warnings.filterwarnings("ignore")
mu.set_options(pull_on_update=False)
sc.settings.verbosity = 0
sc.set_figure_params(
    dpi=80,
    facecolor="white",
    frameon=False,
)

import lamindb as ln

ln.track()
```

--------------------------------

### Install Scanpy using Conda

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/introduction/fundamental_data_structures_and_frameworks.ipynb

Use this command to install the scanpy package via Conda.

```bash
conda install -c conda-forge scanpy
```

--------------------------------

### Build the Book

Source: https://github.com/theislab/single-cell-best-practices/blob/main/CONTRIBUTING.md

Execute this command to build the complete book. This command does not run any notebooks.

```bash
make
```

--------------------------------

### Import Libraries and Initialize LAMINDB Tracking

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/surface_protein/normalization.ipynb

Imports necessary libraries for data manipulation and analysis, and initializes LAMINDB for tracking experiments and artifacts. This setup is standard for data science workflows involving LAMINDB.

```python
import warnings

import matplotlib.pyplot as plt
import muon as mu
import pandas as pd
import scanpy as sc
import seaborn as sns

warnings.filterwarnings("ignore")
mu.set_options(pull_on_update=False)
sc.settings.verbosity = 0
sc.set_figure_params(dpi=80, facecolor="white", frameon=False)

import lamindb as ln

ln.track()
```

--------------------------------

### Install towncrier for changelog management

Source: https://github.com/theislab/single-cell-best-practices/blob/main/CONTRIBUTING.md

Install the towncrier package, used for managing changelog entries for pull requests.

```bash
pip install towncrier
```

--------------------------------

### Initialize Network Visualization Libraries

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/chromatin_accessibility/gene_regulatory_networks_atac.ipynb

Load necessary R libraries for creating network visualizations, including networkD3, r2d3, and imager.

```r
library(networkD3)
library(r2d3)
library(imager)
```

--------------------------------

### Load FigR Library

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/chromatin_accessibility/gene_regulatory_networks_atac.ipynb

Loads the installed FigR package into the current R session. Ensure FigR is installed before running this.

```r
suppressMessages(library(FigR))
```

--------------------------------

### Instantiate VanillaGreedySolver

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/trajectories/lineage_tracing.ipynb

Creates an instance of the VanillaGreedySolver, a heuristic-based algorithm suitable for initial lineage reconstruction.

```python
greedy_solver = cas.solver.VanillaGreedySolver()
```

--------------------------------

### Install Squidpy using pip

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/introduction/advanced_data_structures_and_frameworks.ipynb

Installs the Squidpy package using pip. This is a common method for adding Python libraries to your environment.

```bash
pip install squidpy
```

--------------------------------

### Install MuData

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/introduction/advanced_data_structures_and_frameworks.ipynb

Install the MuData library using pip or conda. This package is used for storing and manipulating multimodal biological data.

```bash
pip install mudata
```

```bash
conda install -c conda-forge mudata
```

--------------------------------

### Setup and Train scVI Model for Batch Correction

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/conditions/compositional.ipynb

Prepares the AnnData object for scVI by setting up the AnnData object with the specified layer and batch key, then trains the scVI model. The number of training epochs is dynamically determined based on the number of observations.

```python
import scvi

adata_scvi = adata[:, adata.var["highly_variable"]].copy()
scvi.model.SCVI.setup_anndata(adata_scvi, layer="counts", batch_key="batch")
model_scvi = scvi.model.SCVI(adata_scvi)
max_epochs_scvi = int(np.min([round((20000 / adata.n_obs) * 400), 400]))
model_scvi.train(max_epochs=max_epochs_scvi)
adata.obsm["X_scVI"] = model_scvi.get_latent_representation()
```

--------------------------------

### Import Libraries and Setup Environment

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/surface_protein/quality_control.ipynb

Imports essential Python libraries for CITE-seq data analysis, including scanpy, muon, and lamindb. It also configures warnings, muon options, scanpy settings, and initializes lamindb tracking.

```python
import warnings

import muon as mu
import numpy as np
import pandas as pd
import scanpy as sc
import seaborn as sns
from scipy.stats import median_abs_deviation

warnings.filterwarnings("ignore")
mu.set_options(pull_on_update=False)
sc.settings.verbosity = 0
sc.set_figure_params(
    dpi=80,
    facecolor="white",
    frameon=False,
)


import lamindb as ln

ln.track()
```

--------------------------------

### Define Data Directory and File Path

Source: https://github.com/theislab/single-cell-best-practices/blob/main/scripts/trajectory/rna_velocity_dataset.ipynb

Sets up the directory for data storage and defines the path for the dataset file.

```python
DATA_DIR = Path("../../data/")
DATA_DIR.mkdir(parents=True, exist_ok=True)

FILE_PATH = DATA_DIR / "pancreas.h5ad"
```

--------------------------------

### Setup AnnData for MultiVAE Model

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/multimodal_integration/advanced_integration.ipynb

Registers categorical covariates for a MultiVAE model. Specifies the end index for RNA data, which is crucial for model setup.

```python
mtg.model.MultiVAE.setup_anndata(
    adata,
    categorical_covariate_keys=["Modality", "Samplename"],
    rna_indices_end=4000,
)
```

--------------------------------

### Install Squidpy using Conda

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/introduction/advanced_data_structures_and_frameworks.ipynb

Installs the Squidpy package using Conda from the conda-forge channel. This is recommended for managing complex dependencies in scientific Python projects.

```bash
conda install -c conda-forge squidpy
```

--------------------------------

### Define Data Paths

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/air_repertoire/ir_profiling.ipynb

Sets up variables for local data storage paths. Ensure the 'data/' directory exists before running.

```python
path_data = "data/"
path_bcr_input = f"{path_data}/BCR_00_read_aligned.csv"
path_tcr_input = f"{path_data}/TCR_00_read_aligned.tsv"
```

--------------------------------

### List All Conda Environments

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/_static/default_text_env_setup.md

Verify the successful creation and activation of your Conda environment by listing all available environments on your system.

```bash
conda env list
```

--------------------------------

### Configure Simpleaf Environment Variables

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/introduction/raw_data_processing.md

Sets the ALEVIN_FRY_HOME environment variable to specify where Simpleaf should store configuration and data. It then runs `simpleaf set-paths` to locate and configure necessary tools.

```bash
# simpleaf needs the environment variable ALEVIN_FRY_HOME to store configuration and data.
# For example, the paths to the underlying programs it uses and the CB permit list
mkdir alevin_fry_home && export ALEVIN_FRY_HOME='alevin_fry_home'

# the simpleaf set-paths command finds the path to the required tools and writes a configuration JSON file in the ALEVIN_FRY_HOME folder.
simpleaf set-paths

```

--------------------------------

### Get AnnData Filename

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/introduction/fundamental_data_structures_and_frameworks.ipynb

Retrieve the filename associated with a backed AnnData object.

```python
adata.filename
```

--------------------------------

### Load a Notebook (Transform)

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/_static/default_text_lamindb_setup.md

Download a specific notebook (Transform) from a LaminDB instance to your current working directory using its URL. Older versions can be accessed by adapting the URL.

```bash
lamin load <notebook url>
```

--------------------------------

### Connect to lamindb instance

Source: https://github.com/theislab/single-cell-best-practices/blob/main/CONTRIBUTING.md

Connect to the 'theislab/sc-best-practices' lamindb instance to manage and access datasets.

```bash
lamin connect theislab/sc-best-practices
```

--------------------------------

### Preliminary Preprocessing and Clustering

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/preprocessing_visualization/normalization.ipynb

Performs initial preprocessing steps including normalization, log transformation, PCA, neighbor computation, and Leiden clustering. This sets up the data for subsequent group-based normalization.

```python
adata_pp = adata.copy()
sc.pp.normalize_total(adata_pp)
sc.pp.log1p(adata_pp)
sc.pp.pca(adata_pp, n_comps=15)
sc.pp.neighbors(adata_pp)
sc.tl.leiden(
    adata_pp, key_added="groups", flavor="igraph", n_iterations=2, directed=False
)
```

--------------------------------

### Initialize LamineDB

Source: https://github.com/theislab/single-cell-best-practices/blob/main/scripts/introduction/interoperability_dataset.ipynb

Imports the lamindb library and asserts the correct instance slug is set. This is a prerequisite for tracking data.

```python
import lamindb as ln
```

```python
assert ln.setup.settings.instance.slug == "theislab/sc-best-practices"
```

--------------------------------

### Get Graph Node and Edge Counts

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/multimodal_integration/advanced_integration.ipynb

Retrieves the total number of nodes and edges in the constructed multimodal graph.

```python
graph.number_of_nodes(), graph.number_of_edges()
```

--------------------------------

### Initialize PyDESeq2 with a Simple Design

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/conditions/differential_gene_expression.ipynb

Construct a PyDESeq2 object with a basic design matrix for initial analysis. This is useful when you want to model the effect of a single factor, like a treatment label, across the entire dataset.

```python
pds2 = pt.tl.PyDESeq2(adata=adata_pb, design="~ label")
```

--------------------------------

### Track Annotation Run

Source: https://github.com/theislab/single-cell-best-practices/blob/main/scripts/cellular_structure/annotation_reference_embedding.ipynb

Starts tracking a new run for annotation. This function should be called at the beginning of a tracking session.

```python
ln.track("fYSEh5kJ5gI8")
```

--------------------------------

### Initialize scArches Model for Query Data

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/cellular_structure/annotation.ipynb

Loads the reference model and prepares it to accept new query data. This step involves specifying the reference model path and freezing dropout for consistency.

```python
scarches_model = sca.models.SCVI.load_query_data(
    adata=adata_to_map_augmented,
    reference_model=str(model_dir),
    freeze_dropout=True,
)
```

--------------------------------

### Load NicheNet Resources

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/mechanisms/cell_cell_communication.ipynb

Loads the NicheNet ligand-target matrix and ligand-receptor network from URLs. Ensure the 'nichenetr' package is installed.

```R
suppressPackageStartupMessages({
    if(!require(nichenetr)) remotes::install_github("saeyslab/nichenetr", upgrade = "never")
})

# Increase timeout threshold
options(timeout=600)

# Load PK
ligand_target_matrix <- readRDS(url("https://zenodo.org/record/7074291/files/ligand_target_matrix_nsga2r_final.rds"))
lr_network <- readRDS(url("https://zenodo.org/record/7074291/files/lr_network_human_21122021.rds"))
```

--------------------------------

### Run helper script for quizzes

Source: https://github.com/theislab/single-cell-best-practices/blob/main/CONTRIBUTING.md

Run the helper script from 'jupyter-book/src/lib.py' to enable quiz and flashcard creation in a notebook.

```python
%run ../src/lib.py
```

--------------------------------

### Load cisTopic Library

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/chromatin_accessibility/gene_regulatory_networks_atac.ipynb

Loads the cisTopic library into the current R session. Use this after installation to access cisTopic functions.

```r
suppressMessages(library(cisTopic))
```

--------------------------------

### Setup R Dependencies with reticulate

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/mechanisms/cell_cell_communication.ipynb

Activates the anndata2ri converter and loads R extensions for use within a Python environment.

```python
# Setting up R dependencies
import anndata2ri

anndata2ri.activate()

%load_ext rpy2.ipython
```

--------------------------------

### Set R Library Path

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/chromatin_accessibility/quality_control.ipynb

Ensures R uses the correct path to the installed environment libraries before running doublet detection.

```R
%%R
.libPaths()
```

--------------------------------

### Create a changelog fragment

Source: https://github.com/theislab/single-cell-best-practices/blob/main/CONTRIBUTING.md

Create a changelog fragment file for a pull request, specifying the change description, PR number, and author.

```bash
towncrier create -c 'update blah blah ([#34](https://github.com/theislab/single-cell-best-practices/pull/34)) <sub>@seohyonkim</sub>' 34.changed.md
```

--------------------------------

### Set R Library Paths

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/chromatin_accessibility/muon_to_seurat.ipynb

Specifies the library paths for R to find installed packages. Ensure these paths are correct for your environment.

```bash
.libPaths(c("/home/icb/laura.martens/miniconda3/envs/best_practices/lib/R/library", "/home/icb/laura.martens/miniconda3/envs/signac/lib/R/library"))
```

--------------------------------

### Initialize and Train totalVI Model

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/multimodal_integration/advanced_integration.ipynb

Initialize the totalVI model with specified architecture parameters and train it using the prepared AnnData object. The training process optimizes the model for multimodal data integration.

```python
arches_params = {
    "use_layer_norm": "both",
    "use_batch_norm": "none",
    "n_layers_decoder": 2,
    "n_layers_encoder": 2,
}

vae = TOTALVI(adata, **arches_params)
vae.train()
```

--------------------------------

### Connect to LaminDB Instance and Load DataFrame

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/_static/default_text_lamindb_setup.md

Connect to a specified LaminDB instance and retrieve the first 100 stored datasets as a pandas DataFrame. Ensure you have authenticated with LaminDB.

```python
import lamindb as ln

ln.Artifact.connect("theislab/sc-best-practices").df()
```

--------------------------------

### Get Latent Representation

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/multimodal_integration/paired_integration.ipynb

Obtains the latent representation from the trained totalVI model and stores it in the AnnData object's 'X_totalVI' attribute.

```python
mdata.obsm["X_totalVI"] = vae.get_latent_representation()
```

--------------------------------

### Initialize totalVI Model

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/multimodal_integration/paired_integration.ipynb

Initializes the totalVI model with the prepared AnnData object. This step may involve computing empirical prior initialization for protein background.

```python
vae = scvi.model.TOTALVI(adata)
```

--------------------------------

### Construct CoNGA Preprocessing Command

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/air_repertoire/multimodal_integration.ipynb

Constructs the command-line string to execute the CoNGA setup script for processing 10x data. This script takes the filtered TCR CSV and outputs a clones file, specifying the organism.

```python
cmd_conga_pp = (
    "python conga/scripts/setup_10x_for_conga.py "
    f"--filtered_contig_annotations_csvfile {path_reduced_tcr} "
    f"--output_clones_file {path_conga_clones} "
    "--organism human"
)

```

--------------------------------

### Load AnnData Object using lamindb

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/preprocessing_visualization/quality_control.ipynb

Loads an AnnData object from a specified artifact using lamindb. Ensure lamindb is installed and configured.

```python
import lamindb as ln

af = ln.Artifact.connect("theislab/sc-best-practices").get(
    key="preprocessing_visualization/quality_control_adata.h5ad", is_latest=True
)
ada = af.load()
ada
```

--------------------------------

### Prepare and Run scCODA Model

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/conditions/compositional.ipynb

Prepares the scCODA data for analysis by specifying the formula and reference cell type, then runs the NUTS sampler for inference. Ensure the 'coda' modality is correctly prepared.

```python
sccoda_data = sccoda_model.prepare(
    sccoda_data,
    modality_key="coda",
    formula="condition",
    reference_cell_type="Endocrine",
)
sccoda_model.run_nuts(sccoda_data, modality_key="coda", rng_key=1234)
```

--------------------------------

### Instantiate CassiopeiaTree

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/trajectories/lineage_tracing.ipynb

Initializes a CassiopeiaTree object with the character matrix and priors. This is a required first step before performing tree inference.

```python
tree = cas.data.CassiopeiaTree(character_matrix=character_matrix, priors=priors)
```

--------------------------------

### Define Cell Types to Check

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/cellular_structure/annotation.ipynb

Creates a list of specific cell types for which marker gene expression will be checked. This is a setup step for further analysis.

```python
cell_types_to_check = [
    "CD14+ Mono",
    "cDC2",
    "NK",
    "B1 B",
    "CD4+ T activated",
    "T naive",
    "MK/E prog",
]
```

--------------------------------

### Quantify Reads with Simpleaf

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/introduction/raw_data_processing.md

Performs quantification using simpleaf quant, mapping FASTQ files to a pre-built index. Options include chemistry type, threads, read files, index path, unspliced permit list, resolution, and output directory.

```bash
simpleaf quant \
-c 10xv3 -t 8 \
-1 $reads1 -2 $reads2 \
-i simpleaf_index/index \
-u -r cr-like \
-m simpleaf_index/index/t2g_3col.tsv \
-o simpleaf_quant
```

--------------------------------

### Load SpatialData Dataset with Lamindb

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/introduction/advanced_data_structures_and_frameworks.ipynb

Loads a SpatialData dataset from a lamindb artifact. This example assumes the artifact 'introduction/advanced_data_structures_and_frameworks_spatialdata.zarr' is available in the lamindb storage.

```python
sdata = ln.Artifact.get(
    key="introduction/advanced_data_structures_and_frameworks_spatialdata.zarr"
).load()
sdata
```

--------------------------------

### Import Libraries and Set Up Scanpy

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/spatial/deconvolution.ipynb

Imports necessary libraries for cell2location, scanpy, and squidpy. Sets verbosity and figure parameters for scanpy.

```python
import cell2location as c2l
import matplotlib
import scanpy as sc
import squidpy as sq

sc.settings.verbosity = 3
sc.settings.set_figure_params(dpi=80, facecolor="white")
```

--------------------------------

### Define Number of Genes and Regions

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/multimodal_integration/paired_integration.ipynb

Defines the number of genes and regions for subsequent model setup. Ensure 'rna' and 'atac' AnnData objects are available.

```python
n_genes = len(rna.var_names)
n_regions = len(atac.var_names)
```

--------------------------------

### Import and Configure CellPhoneDB Method in Liana

Source: https://github.com/theislab/single-cell-best-practices/blob/main/jupyter-book/mechanisms/cell_cell_communication.ipynb

Demonstrates how to import the cellphonedb method using Liana, specifying source and target cell types, and applying filters and ordering based on p-values and expression means. Useful for identifying and prioritizing significant cell-cell interactions.

```python
liana.cellphonedb(
    source_labels=["CD4 T cells", "B cells", "FCGR3A+ Monocytes"],
    target_labels=["CD8 T cells", "CD14+ Monocytes", "NK cells"],
    filter_fun=lambda x: x["cellphone_pvals"] <= 0.01,
    orderby="lr_means",
    orderby_ascending=False,
    top_n=20,
    figure_size=(9, 5),
    size_range=(1, 6),
)
```

--------------------------------

### Track run and load instance settings

Source: https://github.com/theislab/single-cell-best-practices/blob/main/scripts/cellular_structure/anndata_clustering.ipynb

Asserts the current lamindb instance settings and tracks a new run with a specific ID. This is useful for reproducibility and logging.

```python
assert ln.setup.settings.instance.slug == "theislab/sc-best-practices"

ln.track("639B7kNp8Fcb")
```