anndata

Name: anndata
Author: LeonChaoX

41installs61forks

How to install anndata?

leonchaox/qinyan-academic-skills/anndata

$npx -y skills add LeonChaoX/qinyan-academic-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	```bash
21	uv pip install anndata
22
23	# With optional dependencies
24	uv pip install anndata[dev,test,doc]
25	```
26
27	## Quick Start
28
29	### Creating an AnnData object
30	```python
31	import anndata as ad
32	import numpy as np
33	import pandas as pd
34
35	# Minimal creation
36	X = np.random.rand(100, 2000) # 100 cells × 2000 genes
37	adata = ad.AnnData(X)
38
39	# With metadata
40	obs = pd.DataFrame({
41	'cell_type': ['T cell', 'B cell'] * 50,
42	'sample': ['A', 'B'] * 50
43	}, index=[f'cell_{i}' for i in range(100)])
44
45	var = pd.DataFrame({
46	'gene_name': [f'Gene_{i}' for i in range(2000)]
47	}, index=[f'ENSG{i:05d}' for i in range(2000)])
48
49	adata = ad.AnnData(X=X, obs=obs, var=var)
50	```
51
52	### Reading data
53	```python
54	# Read h5ad file
55	adata = ad.read_h5ad('data.h5ad')
56
57	# Read with backed mode (for large files)
58	adata = ad.read_h5ad('large_data.h5ad', backed='r')
59
60	# Read other formats
61	adata = ad.read_csv('data.csv')
62	adata = ad.read_loom('data.loom')
63	adata = ad.read_10x_h5('filtered_feature_bc_matrix.h5')
64	```
65
66	### Writing data
67	```python
68	# Write h5ad file
69	adata.write_h5ad('output.h5ad')
70
71	# Write with compression
72	adata.write_h5ad('output.h5ad', compression='gzip')
73
74	# Write other formats
75	adata.write_zarr('output.zarr')
76	adata.write_csvs('output_dir/')
77	```
78
79	### Basic operations
80	```python
81	# Subset by conditions
82	t_cells = adata[adata.obs['cell_type'] == 'T cell']
83
84	# Subset by indices
85	subset = adata[0:50, 0:100]
86
87	# Add metadata
88	adata.obs['quality_score'] = np.random.rand(adata.n_obs)
89	adata.var['highly_variable'] = np.random.rand(adata.n_vars) > 0.8
90
91	# Access dimensions
92	print(f"{adata.n_obs} observations × {adata.n_vars} variables")
93	```
94
95	## Core Capabilities
96
97	### 1. Data Structure
98
99	Understand the AnnData object structure including X, obs, var, layers, obsm, varm, obsp, varp, uns, and raw components.
100
101	See: `references/data_structure.md` for comprehensive information on:
102	- Core components (X, obs, var, layers, obsm, varm, obsp, varp, uns, raw)
103	- Creating AnnData objects from various sources
104	- Accessing and manipulating data components
105	- Memory-efficient practices
106
107	### 2. Input/Output Operations
108
109	Read and write data in various formats with support for compression, backed mode, and cloud storage.
110
111	See: `references/io_operations.md` for details on:
112	- Native formats (h5ad, zarr)
113	- Alternative formats (CSV, MTX, Loom, 10X, Excel)
114	- Backed mode for large datasets
115	- Remote data access
116	- Format conversion
117	- Performance optimization
118
119	Common commands:
120	```python
121	# Read/write h5ad
122	adata = ad.read_h5ad('data.h5ad', backed='r')
123	adata.write_h5ad('output.h5ad', compression='gzip')
124
125	# Read 10X data
126	adata = ad.read_10x_h5('filtered_feature_bc_matrix.h5')
127
128	# Read MTX format
129	adata = ad.read_mtx('matrix.mtx').T
130	```
131
132	### 3. Concatenation
133
134	Combine multiple AnnData objects along observations or variables with flexible join strategies.
135
136	See: `references/concatenation.md` for comprehensive coverage of:
137	- Basic concatenation (axis=0 for observations, axis=1 for variables)
138	- Join types (inner, outer)
139	- Merge strategies (same, unique, first, only)
140	- Tracking data sources with labels
141	- Lazy concatenation (AnnCollection)
142	- On-disk concatenation for large datasets
143
144	Common commands:
145	```python
146	# Concatenate observations (combine samples)
147	adata = ad.concat(
148	[adata1, adata2, adata3],
149	axis=0,
150	join='inner',
151	label='batch',
152	keys=['batch1', 'batch2', 'batch3']
153	)
154
155	# Concatenate variables (combine modalities)
156	adata = ad.concat([adata_rna, adata_protein], axis=1)
157
158	# Lazy concatenation
159	from anndata.experimental import AnnCollection
160	collection = AnnCollection(
161	['data1.h5ad', 'data2.h5ad'],
162	join_obs='outer',
163	label='dataset'
164	)
165	```
166
167	### 4. Data Manipulation
168
169	Transform, subset, filter, and reorganiz

People who liked this also viewed

6 picks

Type

anndata

41installs61forks

View on GitHub

How to install anndata?

leonchaox/qinyan-academic-skills/anndata

$npx -y skills add LeonChaoX/qinyan-academic-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	```bash
21	uv pip install anndata
22
23	# With optional dependencies
24	uv pip install anndata[dev,test,doc]
25	```
26
27	## Quick Start
28
29	### Creating an AnnData object
30	```python
31	import anndata as ad
32	import numpy as np
33	import pandas as pd
34
35	# Minimal creation
36	X = np.random.rand(100, 2000) # 100 cells × 2000 genes
37	adata = ad.AnnData(X)
38
39	# With metadata
40	obs = pd.DataFrame({
41	'cell_type': ['T cell', 'B cell'] * 50,
42	'sample': ['A', 'B'] * 50
43	}, index=[f'cell_{i}' for i in range(100)])
44
45	var = pd.DataFrame({
46	'gene_name': [f'Gene_{i}' for i in range(2000)]
47	}, index=[f'ENSG{i:05d}' for i in range(2000)])
48
49	adata = ad.AnnData(X=X, obs=obs, var=var)
50	```
51
52	### Reading data
53	```python
54	# Read h5ad file
55	adata = ad.read_h5ad('data.h5ad')
56
57	# Read with backed mode (for large files)
58	adata = ad.read_h5ad('large_data.h5ad', backed='r')
59
60	# Read other formats
61	adata = ad.read_csv('data.csv')
62	adata = ad.read_loom('data.loom')
63	adata = ad.read_10x_h5('filtered_feature_bc_matrix.h5')
64	```
65
66	### Writing data
67	```python
68	# Write h5ad file
69	adata.write_h5ad('output.h5ad')
70
71	# Write with compression
72	adata.write_h5ad('output.h5ad', compression='gzip')
73
74	# Write other formats
75	adata.write_zarr('output.zarr')
76	adata.write_csvs('output_dir/')
77	```
78
79	### Basic operations
80	```python
81	# Subset by conditions
82	t_cells = adata[adata.obs['cell_type'] == 'T cell']
83
84	# Subset by indices
85	subset = adata[0:50, 0:100]
86
87	# Add metadata
88	adata.obs['quality_score'] = np.random.rand(adata.n_obs)
89	adata.var['highly_variable'] = np.random.rand(adata.n_vars) > 0.8
90
91	# Access dimensions
92	print(f"{adata.n_obs} observations × {adata.n_vars} variables")
93	```
94
95	## Core Capabilities
96
97	### 1. Data Structure
98
99	Understand the AnnData object structure including X, obs, var, layers, obsm, varm, obsp, varp, uns, and raw components.
100
101	See: `references/data_structure.md` for comprehensive information on:
102	- Core components (X, obs, var, layers, obsm, varm, obsp, varp, uns, raw)
103	- Creating AnnData objects from various sources
104	- Accessing and manipulating data components
105	- Memory-efficient practices
106
107	### 2. Input/Output Operations
108
109	Read and write data in various formats with support for compression, backed mode, and cloud storage.
110
111	See: `references/io_operations.md` for details on:
112	- Native formats (h5ad, zarr)
113	- Alternative formats (CSV, MTX, Loom, 10X, Excel)
114	- Backed mode for large datasets
115	- Remote data access
116	- Format conversion
117	- Performance optimization
118
119	Common commands:
120	```python
121	# Read/write h5ad
122	adata = ad.read_h5ad('data.h5ad', backed='r')
123	adata.write_h5ad('output.h5ad', compression='gzip')
124
125	# Read 10X data
126	adata = ad.read_10x_h5('filtered_feature_bc_matrix.h5')
127
128	# Read MTX format
129	adata = ad.read_mtx('matrix.mtx').T
130	```
131
132	### 3. Concatenation
133
134	Combine multiple AnnData objects along observations or variables with flexible join strategies.
135
136	See: `references/concatenation.md` for comprehensive coverage of:
137	- Basic concatenation (axis=0 for observations, axis=1 for variables)
138	- Join types (inner, outer)
139	- Merge strategies (same, unique, first, only)
140	- Tracking data sources with labels
141	- Lazy concatenation (AnnCollection)
142	- On-disk concatenation for large datasets
143
144	Common commands:
145	```python
146	# Concatenate observations (combine samples)
147	adata = ad.concat(
148	[adata1, adata2, adata3],
149	axis=0,
150	join='inner',
151	label='batch',
152	keys=['batch1', 'batch2', 'batch3']
153	)
154
155	# Concatenate variables (combine modalities)
156	adata = ad.concat([adata_rna, adata_protein], axis=1)
157
158	# Lazy concatenation
159	from anndata.experimental import AnnCollection
160	collection = AnnCollection(
161	['data1.h5ad', 'data2.h5ad'],
162	join_obs='outer',
163	label='dataset'
164	)
165	```
166
167	### 4. Data Manipulation
168
169	Transform, subset, filter, and reorganiz

People who liked this also viewed

6 picks

Type