arboreto

Name: arboreto
Author: LeonChaoX

42installs61forks

How to install arboreto?

leonchaox/qinyan-academic-skills/arboreto

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

Installs into the current project.

Summary

Infer gene regulatory networks (GRNs) from gene expression data using scalable algorithms (GRNBoost2, GENIE3). Use when analyzing transcriptomics data (bulk RNA-seq, single-cell RNA-seq) to identify transcription factor-target gene relationships and regulatory interactions. Suppo

Files · 1

View on GitHub

SKILL.md

1	# Arboreto
2
3	## Overview
4
5	Arboreto is a computational library for inferring gene regulatory networks (GRNs) from gene expression data using parallelized algorithms that scale from single machines to multi-node clusters.
6
7	Core capability: Identify which transcription factors (TFs) regulate which target genes based on expression patterns across observations (cells, samples, conditions).
8
9	## Quick Start
10
11	Install arboreto:
12	```bash
13	uv pip install arboreto
14	```
15
16	Basic GRN inference:
17	```python
18	import pandas as pd
19	from arboreto.algo import grnboost2
20
21	if __name__ == '__main__':
22	# Load expression data (genes as columns)
23	expression_matrix = pd.read_csv('expression_data.tsv', sep='\t')
24
25	# Infer regulatory network
26	network = grnboost2(expression_data=expression_matrix)
27
28	# Save results (TF, target, importance)
29	network.to_csv('network.tsv', sep='\t', index=False, header=False)
30	```
31
32	Critical: Always use `if __name__ == '__main__':` guard because Dask spawns new processes.
33
34	## Core Capabilities
35
36	### 1. Basic GRN Inference
37
38	For standard GRN inference workflows including:
39	- Input data preparation (Pandas DataFrame or NumPy array)
40	- Running inference with GRNBoost2 or GENIE3
41	- Filtering by transcription factors
42	- Output format and interpretation
43
44	See: `references/basic_inference.md`
45
46	Use the ready-to-run script: `scripts/basic_grn_inference.py` for standard inference tasks:
47	```bash
48	python scripts/basic_grn_inference.py expression_data.tsv output_network.tsv --tf-file tfs.txt --seed 777
49	```
50
51	### 2. Algorithm Selection
52
53	Arboreto provides two algorithms:
54
55	GRNBoost2 (Recommended):
56	- Fast gradient boosting-based inference
57	- Optimized for large datasets (10k+ observations)
58	- Default choice for most analyses
59
60	GENIE3:
61	- Random Forest-based inference
62	- Original multiple regression approach
63	- Use for comparison or validation
64
65	Quick comparison:
66	```python
67	from arboreto.algo import grnboost2, genie3
68
69	# Fast, recommended
70	network_grnboost = grnboost2(expression_data=matrix)
71
72	# Classic algorithm
73	network_genie3 = genie3(expression_data=matrix)
74	```
75
76	For detailed algorithm comparison, parameters, and selection guidance: `references/algorithms.md`
77
78	### 3. Distributed Computing
79
80	Scale inference from local multi-core to cluster environments:
81
82	Local (default) - Uses all available cores automatically:
83	```python
84	network = grnboost2(expression_data=matrix)
85	```
86
87	Custom local client - Control resources:
88	```python
89	from distributed import LocalCluster, Client
90
91	local_cluster = LocalCluster(n_workers=10, memory_limit='8GB')
92	client = Client(local_cluster)
93
94	network = grnboost2(expression_data=matrix, client_or_address=client)
95
96	client.close()
97	local_cluster.close()
98	```
99
100	Cluster computing - Connect to remote Dask scheduler:
101	```python
102	from distributed import Client
103
104	client = Client('tcp://scheduler:8786')
105	network = grnboost2(expression_data=matrix, client_or_address=client)
106	```
107
108	For cluster setup, performance optimization, and large-scale workflows: `references/distributed_computing.md`
109
110	## Installation
111
112	```bash
113	uv pip install arboreto
114	```
115
116	Dependencies: scipy, scikit-learn, numpy, pandas, dask, distributed
117
118	## Common Use Cases
119
120	### Single-Cell RNA-seq Analysis
121	```python
122	import pandas as pd
123	from arboreto.algo import grnboost2
124
125	if __name__ == '__main__':
126	# Load single-cell expression matrix (cells x genes)
127	sc_data = pd.read_csv('scrna_counts.tsv', sep='\t')
128
129	# Infer cell-type-specific regulatory network
130	network = grnboost2(expression_data=sc_data, seed=42)
131
132	# Filter high-confidence links
133	high_confidence = network[network['importance'] > 0.5]
134	high_confidence.to_csv('grn_high_confidence.tsv', sep='\t', index=False)
135	```
136
137	### Bulk RNA-seq with TF Filtering
138	```python
139	from arboreto.utils import load_tf_names
140	from arboreto.algo import grnboost2
141
142	if __name__ == '__main__':
143	# Load data
144	expression_data = pd.read_csv('rnaseq_tpm.tsv', sep='\t')
145	tf_names = load_tf_names('human_tfs.txt')
146
147	# Infer with TF restriction
148	network = grnboost2(
149	expression_data=expression_data,
150	tf_names=tf_names,
151	seed=123
152	)
153
154	network.to_csv('tf_target_network.tsv', sep='\t', index=False)
155	```
156
157	### Comparative Analysis (Multiple Conditions)
158	```python
159	from arboreto.algo import grnboost2
160
161	if __name__ == '__main__':
162	# Infer networks for different conditions
163	conditions = ['control', 'treatment_24h', 'treatment_48h']
164
165	for condition in conditions:
166	data = pd.read_csv(f'{condition}_expression.tsv', sep='\t')

People who liked this also viewed

6 picks

Type

arboreto

42installs61forks

View on GitHub

How to install arboreto?

leonchaox/qinyan-academic-skills/arboreto

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

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Arboreto
2
3	## Overview
4
5	Arboreto is a computational library for inferring gene regulatory networks (GRNs) from gene expression data using parallelized algorithms that scale from single machines to multi-node clusters.
6
7	Core capability: Identify which transcription factors (TFs) regulate which target genes based on expression patterns across observations (cells, samples, conditions).
8
9	## Quick Start
10
11	Install arboreto:
12	```bash
13	uv pip install arboreto
14	```
15
16	Basic GRN inference:
17	```python
18	import pandas as pd
19	from arboreto.algo import grnboost2
20
21	if __name__ == '__main__':
22	# Load expression data (genes as columns)
23	expression_matrix = pd.read_csv('expression_data.tsv', sep='\t')
24
25	# Infer regulatory network
26	network = grnboost2(expression_data=expression_matrix)
27
28	# Save results (TF, target, importance)
29	network.to_csv('network.tsv', sep='\t', index=False, header=False)
30	```
31
32	Critical: Always use `if __name__ == '__main__':` guard because Dask spawns new processes.
33
34	## Core Capabilities
35
36	### 1. Basic GRN Inference
37
38	For standard GRN inference workflows including:
39	- Input data preparation (Pandas DataFrame or NumPy array)
40	- Running inference with GRNBoost2 or GENIE3
41	- Filtering by transcription factors
42	- Output format and interpretation
43
44	See: `references/basic_inference.md`
45
46	Use the ready-to-run script: `scripts/basic_grn_inference.py` for standard inference tasks:
47	```bash
48	python scripts/basic_grn_inference.py expression_data.tsv output_network.tsv --tf-file tfs.txt --seed 777
49	```
50
51	### 2. Algorithm Selection
52
53	Arboreto provides two algorithms:
54
55	GRNBoost2 (Recommended):
56	- Fast gradient boosting-based inference
57	- Optimized for large datasets (10k+ observations)
58	- Default choice for most analyses
59
60	GENIE3:
61	- Random Forest-based inference
62	- Original multiple regression approach
63	- Use for comparison or validation
64
65	Quick comparison:
66	```python
67	from arboreto.algo import grnboost2, genie3
68
69	# Fast, recommended
70	network_grnboost = grnboost2(expression_data=matrix)
71
72	# Classic algorithm
73	network_genie3 = genie3(expression_data=matrix)
74	```
75
76	For detailed algorithm comparison, parameters, and selection guidance: `references/algorithms.md`
77
78	### 3. Distributed Computing
79
80	Scale inference from local multi-core to cluster environments:
81
82	Local (default) - Uses all available cores automatically:
83	```python
84	network = grnboost2(expression_data=matrix)
85	```
86
87	Custom local client - Control resources:
88	```python
89	from distributed import LocalCluster, Client
90
91	local_cluster = LocalCluster(n_workers=10, memory_limit='8GB')
92	client = Client(local_cluster)
93
94	network = grnboost2(expression_data=matrix, client_or_address=client)
95
96	client.close()
97	local_cluster.close()
98	```
99
100	Cluster computing - Connect to remote Dask scheduler:
101	```python
102	from distributed import Client
103
104	client = Client('tcp://scheduler:8786')
105	network = grnboost2(expression_data=matrix, client_or_address=client)
106	```
107
108	For cluster setup, performance optimization, and large-scale workflows: `references/distributed_computing.md`
109
110	## Installation
111
112	```bash
113	uv pip install arboreto
114	```
115
116	Dependencies: scipy, scikit-learn, numpy, pandas, dask, distributed
117
118	## Common Use Cases
119
120	### Single-Cell RNA-seq Analysis
121	```python
122	import pandas as pd
123	from arboreto.algo import grnboost2
124
125	if __name__ == '__main__':
126	# Load single-cell expression matrix (cells x genes)
127	sc_data = pd.read_csv('scrna_counts.tsv', sep='\t')
128
129	# Infer cell-type-specific regulatory network
130	network = grnboost2(expression_data=sc_data, seed=42)
131
132	# Filter high-confidence links
133	high_confidence = network[network['importance'] > 0.5]
134	high_confidence.to_csv('grn_high_confidence.tsv', sep='\t', index=False)
135	```
136
137	### Bulk RNA-seq with TF Filtering
138	```python
139	from arboreto.utils import load_tf_names
140	from arboreto.algo import grnboost2
141
142	if __name__ == '__main__':
143	# Load data
144	expression_data = pd.read_csv('rnaseq_tpm.tsv', sep='\t')
145	tf_names = load_tf_names('human_tfs.txt')
146
147	# Infer with TF restriction
148	network = grnboost2(
149	expression_data=expression_data,
150	tf_names=tf_names,
151	seed=123
152	)
153
154	network.to_csv('tf_target_network.tsv', sep='\t', index=False)
155	```
156
157	### Comparative Analysis (Multiple Conditions)
158	```python
159	from arboreto.algo import grnboost2
160
161	if __name__ == '__main__':
162	# Infer networks for different conditions
163	conditions = ['control', 'treatment_24h', 'treatment_48h']
164
165	for condition in conditions:
166	data = pd.read_csv(f'{condition}_expression.tsv', sep='\t')

People who liked this also viewed

6 picks

Type