anndata

Name: anndata
Author: K-Dense-AI

824installs3.1kforks

How to install anndata?

k-dense-ai/scientific-agent-skills/anndata

$npx -y skills add K-Dense-AI/scientific-agent-skills --skill anndata

Installs into the current project.

Summary

Data structure for annotated matrices in single-cell analysis. Use when working with .h5ad files or integrating with the scverse ecosystem. This is the data format skill—for analysis workflows use scanpy; for probabilistic models use scvi-tools; for population-scale queries use c

Files · 1

View on GitHub

SKILL.md

1	# AnnData
2
3	## Overview
4
5	AnnData is a Python package for handling annotated data matrices, storing experimental measurements (X) alongside observation metadata (obs), variable metadata (var), and multi-dimensional annotations (obsm, varm, obsp, varp, uns). Originally designed for single-cell genomics through Scanpy, it now serves as a general-purpose framework for any annotated data requiring efficient storage, manipulation, and analysis.
6
7	## When to Use This Skill
8
9	Use this skill when:
10	- Creating, reading, or writing AnnData objects
11	- Working with h5ad, zarr, or other genomics data formats
12	- Performing single-cell RNA-seq analysis
13	- Managing large datasets with sparse matrices or backed mode
14	- Concatenating multiple datasets or experimental batches
15	- Subsetting, filtering, or transforming annotated data
16	- Integrating with scanpy, scvi-tools, or other scverse ecosystem tools
17
18	## Installation
19
20	Requires Python 3.11+. Current stable release: 0.12.16 (released 2026-05-18).
21
22	```bash
23	uv pip install "anndata==0.12.16"
24
25	# Lazy I/O and dask-backed operations
26	uv pip install "anndata[dask,lazy]==0.12.16"
27
28	# Development / docs (contributors)
29	uv pip install "anndata[dev,test,doc]==0.12.16"
30	```
31
32	Use unpinned installs only when intentionally tracking the latest compatible release.
33
34	Current API notes:
35	- Use `anndata.io` for non-native `read_` and `write_` helpers. Top-level `anndata.read_h5ad` and `anndata.read_zarr` remain supported.
36	- Avoid deprecated APIs: `ad.read`, `AnnData.concatenate()`, `AnnData.*_keys()`, and `anndata.__version__`. Prefer `ad.read_h5ad`, `ad.concat`, mapping `.keys()`, and `importlib.metadata.version("anndata")`.
37	- Treat `anndata.experimental` APIs as useful but unstable. Prefer them for large-data workflows only when their current caveats are acceptable.
38
39	## Quick Start
40
41	### Creating an AnnData object
42	```python
43	import anndata as ad
44	import numpy as np
45	import pandas as pd
46
47	# Minimal creation
48	X = np.random.rand(100, 2000) # 100 cells × 2000 genes
49	adata = ad.AnnData(X)
50
51	# With metadata
52	obs = pd.DataFrame({
53	'cell_type': ['T cell', 'B cell'] * 50,
54	'sample': ['A', 'B'] * 50
55	}, index=[f'cell_{i}' for i in range(100)])
56
57	var = pd.DataFrame({
58	'gene_name': [f'Gene_{i}' for i in range(2000)]
59	}, index=[f'ENSG{i:05d}' for i in range(2000)])
60
61	adata = ad.AnnData(X=X, obs=obs, var=var)
62	```
63
64	### Reading data
65	```python
66	# Native formats (read_h5ad/read_zarr remain at top-level)
67	adata = ad.read_h5ad('data.h5ad')
68	adata = ad.read_h5ad('large_data.h5ad', backed='r') # lazy load for large files
69	adata = ad.read_zarr('data.zarr')
70
71	# Other formats: prefer anndata.io (top-level imports are deprecated)
72	from anndata.io import read_csv, read_loom, read_mtx
73
74	adata = read_csv('data.csv')
75	adata = read_loom('data.loom')
76
77	# 10X Genomics: use scanpy (not anndata) — see scanpy skill
78	import scanpy as sc
79	adata = sc.read_10x_h5('filtered_feature_bc_matrix.h5')
80	adata = sc.read_10x_mtx('filtered_feature_bc_matrix/')
81	```
82
83	### Writing data
84	```python
85	# Write h5ad file
86	adata.write_h5ad('output.h5ad')
87
88	# Write with compression
89	adata.write_h5ad('output.h5ad', compression='gzip')
90
91	# Write other formats
92	adata.write_zarr('output.zarr')
93	adata.write_csvs('output_dir/')
94	```
95
96	### Basic operations
97	```python
98	# Subset by conditions
99	t_cells = adata[adata.obs['cell_type'] == 'T cell']
100
101	# Subset by indices
102	subset = adata[0:50, 0:100]
103
104	# Add metadata
105	adata.obs['quality_score'] = np.random.rand(adata.n_obs)
106	adata.var['highly_variable'] = np.random.rand(adata.n_vars) > 0.8
107
108	# Access dimensions
109	print(f"{adata.n_obs} observations × {adata.n_vars} variables")
110	```
111
112	## Core Capabilities
113
114	### 1. Data Structure
115
116	Understand the AnnData object structure including X, obs, var, layers, obsm, varm, obsp, varp, uns, and raw components.
117
118	See: `references/data_structure.md` for comprehensive information on:
119	- Core components (X, obs, var, layers, obsm, varm, obsp, varp, uns, raw)
120	- Creating AnnData objects from various sources
121	- Accessing and manipulating data components
122	- Memory-efficient practices
123
124	### 2. Input/Output Operations
125
126	Read and write data in various formats with support for compression, backed mode, and cloud storage.
127
128	See: `references/io_operations.md` for details on:
129	- Native formats (h5ad, zarr)
130	- Alternative formats (CSV, MTX, Loom, 10X, Excel)
131	- Backed mode for large datasets
132	- Remote data access
133	- Format conversion
134	- Performance optimization
135
136	Common commands:

People who liked this also viewed

6 picks

Type

anndata

824installs3.1kforks

View on GitHub

How to install anndata?

k-dense-ai/scientific-agent-skills/anndata

$npx -y skills add K-Dense-AI/scientific-agent-skills --skill anndata

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# AnnData
2
3	## Overview
4
5	AnnData is a Python package for handling annotated data matrices, storing experimental measurements (X) alongside observation metadata (obs), variable metadata (var), and multi-dimensional annotations (obsm, varm, obsp, varp, uns). Originally designed for single-cell genomics through Scanpy, it now serves as a general-purpose framework for any annotated data requiring efficient storage, manipulation, and analysis.
6
7	## When to Use This Skill
8
9	Use this skill when:
10	- Creating, reading, or writing AnnData objects
11	- Working with h5ad, zarr, or other genomics data formats
12	- Performing single-cell RNA-seq analysis
13	- Managing large datasets with sparse matrices or backed mode
14	- Concatenating multiple datasets or experimental batches
15	- Subsetting, filtering, or transforming annotated data
16	- Integrating with scanpy, scvi-tools, or other scverse ecosystem tools
17
18	## Installation
19
20	Requires Python 3.11+. Current stable release: 0.12.16 (released 2026-05-18).
21
22	```bash
23	uv pip install "anndata==0.12.16"
24
25	# Lazy I/O and dask-backed operations
26	uv pip install "anndata[dask,lazy]==0.12.16"
27
28	# Development / docs (contributors)
29	uv pip install "anndata[dev,test,doc]==0.12.16"
30	```
31
32	Use unpinned installs only when intentionally tracking the latest compatible release.
33
34	Current API notes:
35	- Use `anndata.io` for non-native `read_` and `write_` helpers. Top-level `anndata.read_h5ad` and `anndata.read_zarr` remain supported.
36	- Avoid deprecated APIs: `ad.read`, `AnnData.concatenate()`, `AnnData.*_keys()`, and `anndata.__version__`. Prefer `ad.read_h5ad`, `ad.concat`, mapping `.keys()`, and `importlib.metadata.version("anndata")`.
37	- Treat `anndata.experimental` APIs as useful but unstable. Prefer them for large-data workflows only when their current caveats are acceptable.
38
39	## Quick Start
40
41	### Creating an AnnData object
42	```python
43	import anndata as ad
44	import numpy as np
45	import pandas as pd
46
47	# Minimal creation
48	X = np.random.rand(100, 2000) # 100 cells × 2000 genes
49	adata = ad.AnnData(X)
50
51	# With metadata
52	obs = pd.DataFrame({
53	'cell_type': ['T cell', 'B cell'] * 50,
54	'sample': ['A', 'B'] * 50
55	}, index=[f'cell_{i}' for i in range(100)])
56
57	var = pd.DataFrame({
58	'gene_name': [f'Gene_{i}' for i in range(2000)]
59	}, index=[f'ENSG{i:05d}' for i in range(2000)])
60
61	adata = ad.AnnData(X=X, obs=obs, var=var)
62	```
63
64	### Reading data
65	```python
66	# Native formats (read_h5ad/read_zarr remain at top-level)
67	adata = ad.read_h5ad('data.h5ad')
68	adata = ad.read_h5ad('large_data.h5ad', backed='r') # lazy load for large files
69	adata = ad.read_zarr('data.zarr')
70
71	# Other formats: prefer anndata.io (top-level imports are deprecated)
72	from anndata.io import read_csv, read_loom, read_mtx
73
74	adata = read_csv('data.csv')
75	adata = read_loom('data.loom')
76
77	# 10X Genomics: use scanpy (not anndata) — see scanpy skill
78	import scanpy as sc
79	adata = sc.read_10x_h5('filtered_feature_bc_matrix.h5')
80	adata = sc.read_10x_mtx('filtered_feature_bc_matrix/')
81	```
82
83	### Writing data
84	```python
85	# Write h5ad file
86	adata.write_h5ad('output.h5ad')
87
88	# Write with compression
89	adata.write_h5ad('output.h5ad', compression='gzip')
90
91	# Write other formats
92	adata.write_zarr('output.zarr')
93	adata.write_csvs('output_dir/')
94	```
95
96	### Basic operations
97	```python
98	# Subset by conditions
99	t_cells = adata[adata.obs['cell_type'] == 'T cell']
100
101	# Subset by indices
102	subset = adata[0:50, 0:100]
103
104	# Add metadata
105	adata.obs['quality_score'] = np.random.rand(adata.n_obs)
106	adata.var['highly_variable'] = np.random.rand(adata.n_vars) > 0.8
107
108	# Access dimensions
109	print(f"{adata.n_obs} observations × {adata.n_vars} variables")
110	```
111
112	## Core Capabilities
113
114	### 1. Data Structure
115
116	Understand the AnnData object structure including X, obs, var, layers, obsm, varm, obsp, varp, uns, and raw components.
117
118	See: `references/data_structure.md` for comprehensive information on:
119	- Core components (X, obs, var, layers, obsm, varm, obsp, varp, uns, raw)
120	- Creating AnnData objects from various sources
121	- Accessing and manipulating data components
122	- Memory-efficient practices
123
124	### 2. Input/Output Operations
125
126	Read and write data in various formats with support for compression, backed mode, and cloud storage.
127
128	See: `references/io_operations.md` for details on:
129	- Native formats (h5ad, zarr)
130	- Alternative formats (CSV, MTX, Loom, 10X, Excel)
131	- Backed mode for large datasets
132	- Remote data access
133	- Format conversion
134	- Performance optimization
135
136	Common commands:

People who liked this also viewed

6 picks

Type