pydeseq2

Name: pydeseq2
Author: LeonChaoX

698stars61forks

How to install pydeseq2?

leonchaox/qinyan-academic-skills/pydeseq2

$npx -y skills add LeonChaoX/qinyan-academic-skills --skill pydeseq2

Installs into the current project.

Summary

Differential gene expression analysis (Python DESeq2). Identify DE genes from bulk RNA-seq counts, Wald tests, FDR correction, volcano/MA plots, for RNA-seq analysis.

Files · 1

View on GitHub

SKILL.md

1	# PyDESeq2
2
3	## Overview
4
5	PyDESeq2 is a Python implementation of DESeq2 for differential expression analysis with bulk RNA-seq data. Design and execute complete workflows from data loading through result interpretation, including single-factor and multi-factor designs, Wald tests with multiple testing correction, optional apeGLM shrinkage, and integration with pandas and AnnData.
6
7	## When to Use This Skill
8
9	This skill should be used when:
10	- Analyzing bulk RNA-seq count data for differential expression
11	- Comparing gene expression between experimental conditions (e.g., treated vs control)
12	- Performing multi-factor designs accounting for batch effects or covariates
13	- Converting R-based DESeq2 workflows to Python
14	- Integrating differential expression analysis into Python-based pipelines
15	- Users mention "DESeq2", "differential expression", "RNA-seq analysis", or "PyDESeq2"
16
17	## Quick Start Workflow
18
19	For users who want to perform a standard differential expression analysis:
20
21	```python
22	import pandas as pd
23	from pydeseq2.dds import DeseqDataSet
24	from pydeseq2.ds import DeseqStats
25
26	# 1. Load data
27	counts_df = pd.read_csv("counts.csv", index_col=0).T # Transpose to samples × genes
28	metadata = pd.read_csv("metadata.csv", index_col=0)
29
30	# 2. Filter low-count genes
31	genes_to_keep = counts_df.columns[counts_df.sum(axis=0) >= 10]
32	counts_df = counts_df[genes_to_keep]
33
34	# 3. Initialize and fit DESeq2
35	dds = DeseqDataSet(
36	counts=counts_df,
37	metadata=metadata,
38	design="~condition",
39	refit_cooks=True
40	)
41	dds.deseq2()
42
43	# 4. Perform statistical testing
44	ds = DeseqStats(dds, contrast=["condition", "treated", "control"])
45	ds.summary()
46
47	# 5. Access results
48	results = ds.results_df
49	significant = results[results.padj < 0.05]
50	print(f"Found {len(significant)} significant genes")
51	```
52
53	## Core Workflow Steps
54
55	### Step 1: Data Preparation
56
57	Input requirements:
58	- Count matrix: Samples × genes DataFrame with non-negative integer read counts
59	- Metadata: Samples × variables DataFrame with experimental factors
60
61	Common data loading patterns:
62
63	```python
64	# From CSV (typical format: genes × samples, needs transpose)
65	counts_df = pd.read_csv("counts.csv", index_col=0).T
66	metadata = pd.read_csv("metadata.csv", index_col=0)
67
68	# From TSV
69	counts_df = pd.read_csv("counts.tsv", sep="\t", index_col=0).T
70
71	# From AnnData
72	import anndata as ad
73	adata = ad.read_h5ad("data.h5ad")
74	counts_df = pd.DataFrame(adata.X, index=adata.obs_names, columns=adata.var_names)
75	metadata = adata.obs
76	```
77
78	Data filtering:
79
80	```python
81	# Remove low-count genes
82	genes_to_keep = counts_df.columns[counts_df.sum(axis=0) >= 10]
83	counts_df = counts_df[genes_to_keep]
84
85	# Remove samples with missing metadata
86	samples_to_keep = ~metadata.condition.isna()
87	counts_df = counts_df.loc[samples_to_keep]
88	metadata = metadata.loc[samples_to_keep]
89	```
90
91	### Step 2: Design Specification
92
93	The design formula specifies how gene expression is modeled.
94
95	Single-factor designs:
96	```python
97	design = "~condition" # Simple two-group comparison
98	```
99
100	Multi-factor designs:
101	```python
102	design = "~batch + condition" # Control for batch effects
103	design = "~age + condition" # Include continuous covariate
104	design = "~group + condition + group:condition" # Interaction effects
105	```
106
107	Design formula guidelines:
108	- Use Wilkinson formula notation (R-style)
109	- Put adjustment variables (e.g., batch) before the main variable of interest
110	- Ensure variables exist as columns in the metadata DataFrame
111	- Use appropriate data types (categorical for discrete variables)
112
113	### Step 3: DESeq2 Fitting
114
115	Initialize the DeseqDataSet and run the complete pipeline:
116
117	```python
118	from pydeseq2.dds import DeseqDataSet
119
120	dds = DeseqDataSet(
121	counts=counts_df,
122	metadata=metadata,
123	design="~condition",
124	refit_cooks=True, # Refit after removing outliers
125	n_cpus=1 # Parallel processing (adjust as needed)
126	)
127
128	# Run the complete DESeq2 pipeline
129	dds.deseq2()
130	```
131
132	What `deseq2()` does:
133	1. Computes size factors (normalization)
134	2. Fits genewise dispersions
135	3. Fits dispersion trend curve
136	4. Computes dispersion priors
137	5. Fits MAP dispersions (shrinkage)
138	6. Fits log fold changes
139	7. Calculates Cook's distances (outlier detection)
140	8. Refits if outliers detected (optional)
141
142	### Step 4: Statistical Testing
143
144	Perform Wald tests to identify differentially expressed genes:
145
146	```python
147	from pydeseq2.ds import DeseqStats
148
149	ds = DeseqStats(
150	dds,
151	contrast=["condition", "treated", "control"], # Test treated vs control
152	alpha=0.05, # Significance threshold
153	cooks_filter=True, # Filter outliers
154	independent_filter=True # Filter low-power tests
155	)
156
157	ds.summary()
158	```
159
160	Contrast specification:
161	- Format: `[variable, test_level, referenc

People who liked this also viewed

6 picks

Type

find-skillsHelps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express interest in extending capabilities. This skSkillsJul 20262.6M27k
grill-meA relentless interview to sharpen a plan or design.SkillsJul 2026633k182k
vercel-react-best-practicesVercel's official collection of agent skillsSkillsJul 2026573k29k
improve-codebase-architectureScan a codebase for deepening opportunities, present them as a visual HTML report, then grill through whichever one you pick.SkillsJul 2026522k182k
setup-matt-pocock-skillsConfigure this repo for the engineering skills — set up its issue tracker, triage label vocabulary, and domain doc layout. Run once before first use of the other engineering skills.SkillsJul 2026454k182k
triageMove issues and external PRs through a state machine of triage roles — categorise, verify, grill if needed, and write agent-ready briefs.SkillsJul 2026424k182k

pydeseq2

698stars61forks

View on GitHub

How to install pydeseq2?

leonchaox/qinyan-academic-skills/pydeseq2

$npx -y skills add LeonChaoX/qinyan-academic-skills --skill pydeseq2

Installs into the current project.

Summary

Differential gene expression analysis (Python DESeq2). Identify DE genes from bulk RNA-seq counts, Wald tests, FDR correction, volcano/MA plots, for RNA-seq analysis.

Files · 1

View on GitHub

SKILL.md

1	# PyDESeq2
2
3	## Overview
4
5	PyDESeq2 is a Python implementation of DESeq2 for differential expression analysis with bulk RNA-seq data. Design and execute complete workflows from data loading through result interpretation, including single-factor and multi-factor designs, Wald tests with multiple testing correction, optional apeGLM shrinkage, and integration with pandas and AnnData.
6
7	## When to Use This Skill
8
9	This skill should be used when:
10	- Analyzing bulk RNA-seq count data for differential expression
11	- Comparing gene expression between experimental conditions (e.g., treated vs control)
12	- Performing multi-factor designs accounting for batch effects or covariates
13	- Converting R-based DESeq2 workflows to Python
14	- Integrating differential expression analysis into Python-based pipelines
15	- Users mention "DESeq2", "differential expression", "RNA-seq analysis", or "PyDESeq2"
16
17	## Quick Start Workflow
18
19	For users who want to perform a standard differential expression analysis:
20
21	```python
22	import pandas as pd
23	from pydeseq2.dds import DeseqDataSet
24	from pydeseq2.ds import DeseqStats
25
26	# 1. Load data
27	counts_df = pd.read_csv("counts.csv", index_col=0).T # Transpose to samples × genes
28	metadata = pd.read_csv("metadata.csv", index_col=0)
29
30	# 2. Filter low-count genes
31	genes_to_keep = counts_df.columns[counts_df.sum(axis=0) >= 10]
32	counts_df = counts_df[genes_to_keep]
33
34	# 3. Initialize and fit DESeq2
35	dds = DeseqDataSet(
36	counts=counts_df,
37	metadata=metadata,
38	design="~condition",
39	refit_cooks=True
40	)
41	dds.deseq2()
42
43	# 4. Perform statistical testing
44	ds = DeseqStats(dds, contrast=["condition", "treated", "control"])
45	ds.summary()
46
47	# 5. Access results
48	results = ds.results_df
49	significant = results[results.padj < 0.05]
50	print(f"Found {len(significant)} significant genes")
51	```
52
53	## Core Workflow Steps
54
55	### Step 1: Data Preparation
56
57	Input requirements:
58	- Count matrix: Samples × genes DataFrame with non-negative integer read counts
59	- Metadata: Samples × variables DataFrame with experimental factors
60
61	Common data loading patterns:
62
63	```python
64	# From CSV (typical format: genes × samples, needs transpose)
65	counts_df = pd.read_csv("counts.csv", index_col=0).T
66	metadata = pd.read_csv("metadata.csv", index_col=0)
67
68	# From TSV
69	counts_df = pd.read_csv("counts.tsv", sep="\t", index_col=0).T
70
71	# From AnnData
72	import anndata as ad
73	adata = ad.read_h5ad("data.h5ad")
74	counts_df = pd.DataFrame(adata.X, index=adata.obs_names, columns=adata.var_names)
75	metadata = adata.obs
76	```
77
78	Data filtering:
79
80	```python
81	# Remove low-count genes
82	genes_to_keep = counts_df.columns[counts_df.sum(axis=0) >= 10]
83	counts_df = counts_df[genes_to_keep]
84
85	# Remove samples with missing metadata
86	samples_to_keep = ~metadata.condition.isna()
87	counts_df = counts_df.loc[samples_to_keep]
88	metadata = metadata.loc[samples_to_keep]
89	```
90
91	### Step 2: Design Specification
92
93	The design formula specifies how gene expression is modeled.
94
95	Single-factor designs:
96	```python
97	design = "~condition" # Simple two-group comparison
98	```
99
100	Multi-factor designs:
101	```python
102	design = "~batch + condition" # Control for batch effects
103	design = "~age + condition" # Include continuous covariate
104	design = "~group + condition + group:condition" # Interaction effects
105	```
106
107	Design formula guidelines:
108	- Use Wilkinson formula notation (R-style)
109	- Put adjustment variables (e.g., batch) before the main variable of interest
110	- Ensure variables exist as columns in the metadata DataFrame
111	- Use appropriate data types (categorical for discrete variables)
112
113	### Step 3: DESeq2 Fitting
114
115	Initialize the DeseqDataSet and run the complete pipeline:
116
117	```python
118	from pydeseq2.dds import DeseqDataSet
119
120	dds = DeseqDataSet(
121	counts=counts_df,
122	metadata=metadata,
123	design="~condition",
124	refit_cooks=True, # Refit after removing outliers
125	n_cpus=1 # Parallel processing (adjust as needed)
126	)
127
128	# Run the complete DESeq2 pipeline
129	dds.deseq2()
130	```
131
132	What `deseq2()` does:
133	1. Computes size factors (normalization)
134	2. Fits genewise dispersions
135	3. Fits dispersion trend curve
136	4. Computes dispersion priors
137	5. Fits MAP dispersions (shrinkage)
138	6. Fits log fold changes
139	7. Calculates Cook's distances (outlier detection)
140	8. Refits if outliers detected (optional)
141
142	### Step 4: Statistical Testing
143
144	Perform Wald tests to identify differentially expressed genes:
145
146	```python
147	from pydeseq2.ds import DeseqStats
148
149	ds = DeseqStats(
150	dds,
151	contrast=["condition", "treated", "control"], # Test treated vs control
152	alpha=0.05, # Significance threshold
153	cooks_filter=True, # Filter outliers
154	independent_filter=True # Filter low-power tests
155	)
156
157	ds.summary()
158	```
159
160	Contrast specification:
161	- Format: `[variable, test_level, referenc

People who liked this also viewed

6 picks

Type