Built-in Datasets

def pbmc3k():
    """
    3k PBMCs from 10x Genomics.
    
    Returns:
    AnnData: 3000 PBMCs with 32738 genes
    """

def pbmc3k_processed():
    """
    Processed 3k PBMCs with cluster annotations.
    
    Returns:
    AnnData: Preprocessed 3k PBMCs with clustering results
    """

def pbmc68k_reduced():
    """
    68k PBMCs from 10x Genomics, reduced for computational efficiency.
    
    Returns:
    AnnData: Subsampled and preprocessed 68k PBMCs
    """

def paul15():
    """
    Hematopoietic stem and progenitor cell dataset from Paul et al. 2015.
    
    Single-cell RNA-seq of 2730 cells from mouse bone marrow capturing
    hematopoietic differentiation. Contains pre-computed diffusion map
    and branch assignments.
    
    Returns:
    AnnData: 2730 cells with 3451 genes and trajectory annotations
    """

def moignard15():
    """
    Blood development dataset from Moignard et al. 2015.
    
    Single-cell qPCR data of 3934 cells with 42 genes during early
    blood development in mouse embryos.
    
    Returns:
    AnnData: 3934 cells with 42 genes and developmental stage annotations
    """

def krumsiek11():
    """
    Krumsiek et al. 2011 hematopoiesis model dataset.
    
    Simulated gene expression data based on Boolean network model
    of hematopoietic differentiation.
    
    Returns:
    AnnData: Simulated cells with continuous gene expression values
    """

def burczynski06():
    """
    Burczynski et al. 2006 Crohn's disease dataset.
    
    Microarray data from colonic mucosal biopsies of Crohn's disease
    patients and healthy controls.
    
    Returns:
    AnnData: 127 samples with 22283 genes and disease annotations
    """

def blobs(n_observations=640, n_variables=11, n_centers=5, cluster_std=1.0, center_box=(-10.0, 10.0), random_state=0):
    """
    Generate synthetic single-cell-like data with Gaussian blobs.
    
    Parameters:
    - n_observations (int): Number of cells to generate
    - n_variables (int): Number of genes
    - n_centers (int): Number of cluster centers
    - cluster_std (float): Standard deviation of clusters
    - center_box (tuple): Bounding box for cluster centers
    - random_state (int): Random seed
    
    Returns:
    AnnData: Synthetic dataset with cluster labels
    """

def toggleswitch(n_observations=640, n_variables=11, random_state=0):
    """
    Toggle switch synthetic dataset.
    
    Simulated data representing a bistable gene regulatory network
    (toggle switch) with two stable states.
    
    Parameters:
    - n_observations (int): Number of cells
    - n_variables (int): Number of genes
    - random_state (int): Random seed
    
    Returns:
    AnnData: Synthetic toggle switch data
    """

def visium_sge(sample='V1_Breast_Cancer_Block_A_Section_1'):
    """
    10x Visium spatial gene expression dataset.
    
    Visium spatial transcriptomics data from various tissue samples
    including breast cancer, mouse brain, and other tissues.
    
    Parameters:
    - sample (str): Sample identifier to load
    
    Returns:
    AnnData: Spatial transcriptomics data with coordinates
    """

def ebi_expression_atlas(accession, filter_cells=True, filter_genes=True):
    """
    Load dataset from EBI Single Cell Expression Atlas.
    
    Parameters:
    - accession (str): EBI Expression Atlas accession number
    - filter_cells (bool): Apply cell filtering
    - filter_genes (bool): Apply gene filtering
    
    Returns:
    AnnData: Dataset from EBI Expression Atlas
    """

import scanpy as sc

# Load 3k PBMCs for tutorial
adata = sc.datasets.pbmc3k()
adata.var_names_unique()

# Load processed PBMCs with annotations
adata = sc.datasets.pbmc3k_processed()
print(adata.obs.columns)  # see available annotations

# Load developmental dataset
adata = sc.datasets.paul15()
print(adata.obs['paul15_clusters'].value_counts())

# Load Visium spatial data
adata = sc.datasets.visium_sge()

# Spatial coordinates are in adata.obsm['spatial']
print(adata.obsm['spatial'].shape)

# Basic spatial plot
sc.pl.spatial(adata, color='total_counts')

# Generate synthetic clustered data
adata = sc.datasets.blobs(n_observations=1000, n_variables=50, n_centers=8)

# Basic analysis
sc.pp.neighbors(adata)
sc.tl.umap(adata)
sc.pl.umap(adata, color='blobs')

# Toggle switch data for trajectory analysis
adata = sc.datasets.toggleswitch()
sc.pp.neighbors(adata)
sc.tl.diffmap(adata)
sc.pl.diffmap(adata, color='distance')

# Paul et al. hematopoiesis data
adata = sc.datasets.paul15()

# The dataset comes with pre-computed results
sc.pl.diffmap(adata, color='paul15_clusters', components=['1,2', '1,3'])

# Moignard et al. early development
adata = sc.datasets.moignard15()
sc.pl.scatter(adata, x='Gata1', y='Gata2', color='groups')

# Crohn's disease dataset
adata = sc.datasets.burczynski06()

# Basic differential expression
sc.tl.rank_genes_groups(adata, 'DISEASE', method='wilcoxon')
sc.pl.rank_genes_groups(adata, n_genes=10)

# Load data from EBI Expression Atlas
# (requires internet connection)
adata = sc.datasets.ebi_expression_atlas('E-MTAB-5061')
print(f"Loaded {adata.n_obs} cells and {adata.n_vars} genes")