rdkit

Name: rdkit
Author: ai4protein

244stars35forks

How to install rdkit?

ai4protein/venusfactory2/rdkit

$npx -y skills add ai4protein/VenusFactory2 --skill rdkit

Installs into the current project.

Summary

Cheminformatics toolkit for fine-grained molecular control. SMILES/SDF parsing, descriptors (MW, LogP, TPSA), fingerprints, substructure search, 2D/3D generation, similarity, reactions. For standard workflows with simpler interface, use datamol (wrapper around RDKit). Use rdkit f

Files · 1

View on GitHub

SKILL.md

1	# RDKit Cheminformatics Toolkit
2
3	## Overview
4
5	RDKit is a comprehensive cheminformatics library providing Python APIs for molecular analysis and manipulation. This skill provides guidance for reading/writing molecular structures, calculating descriptors, fingerprinting, substructure searching, chemical reactions, 2D/3D coordinate generation, and molecular visualization. Use this skill for drug discovery, computational chemistry, and cheminformatics research tasks.
6
7	## Quick Start and Project Scripts
8
9	Scripts live in `src/tools/bioinfo/rdkit/`. Each module is an executable atomic script with `if __name__ == "__main__"` (CLI via argparse).
10
11	\| Script \| Purpose \| Main \|
12	\|--------\|---------\|------\|
13	\| `molecular_properties.py` \| Properties: MW, LogP, TPSA, Lipinski, QED; single SMILES or batch file → CSV \| Yes (argparse) \|
14	\| `substructure_filter.py` \| Filter by SMARTS/SMILES include/exclude; pattern libs (functional-groups, rings, pains, privileged) \| Yes (argparse) \|
15	\| `similarity_search.py` \| Fingerprint similarity (Morgan, RDKit, MACCS, atompair, torsion); Tanimoto/Dice/cosine \| Yes (argparse) \|
16
17	Import from package:
18	```python
19	from src.tools.bioinfo.rdkit import (
20	calculate_properties,
21	process_single_molecule,
22	process_file,
23	filter_molecules,
24	create_pattern_query,
25	PATTERN_LIBRARIES,
26	generate_fingerprint,
27	similarity_search,
28	FINGERPRINT_METHODS,
29	)
30	# Example: single-molecule properties
31	props = calculate_properties(Chem.MolFromSmiles("CCO"))
32	# Example: substructure filter (mols from load_molecules_filter)
33	from src.tools.bioinfo.rdkit.substructure_filter import load_molecules as load_mols
34	mols = load_mols("molecules.smi")
35	filtered, info = filter_molecules(mols, include_patterns=[("benzene", Chem.MolFromSmarts("c1ccccc1"))])
36	# Example: similarity (query_mol + list of dicts with 'mol', 'smiles', 'name', 'index')
37	from src.tools.bioinfo.rdkit.similarity_search import load_molecules as load_db
38	db = load_db("database.smi")
39	hits = similarity_search(query_mol, db, method="morgan", threshold=0.7)
40	```
41
42	Run as CLI (each has `main()`):
43	```bash
44	python -m src.tools.bioinfo.rdkit.molecular_properties "CCO"
45	python -m src.tools.bioinfo.rdkit.molecular_properties --file molecules.smi --output props.csv
46	python -m src.tools.bioinfo.rdkit.substructure_filter molecules.smi --pattern "c1ccccc1" -o filtered.smi
47	python -m src.tools.bioinfo.rdkit.similarity_search "CCO" database.smi --threshold 0.7 -o hits.csv
48	```
49
50	References: `references/api_reference.md`, `references/descriptors_reference.md`, `references/smarts_patterns.md`.
51
52	## Core Capabilities
53
54	### 1. Molecular I/O and Creation
55
56	Reading Molecules:
57
58	Read molecular structures from various formats:
59
60	```python
61	from rdkit import Chem
62
63	# From SMILES strings
64	mol = Chem.MolFromSmiles('Cc1ccccc1') # Returns Mol object or None
65
66	# From MOL files
67	mol = Chem.MolFromMolFile('path/to/file.mol')
68
69	# From MOL blocks (string data)
70	mol = Chem.MolFromMolBlock(mol_block_string)
71
72	# From InChI
73	mol = Chem.MolFromInchi('InChI=1S/C6H6/c1-2-4-6-5-3-1/h1-6H')
74	```
75
76	Writing Molecules:
77
78	Convert molecules to text representations:
79
80	```python
81	# To canonical SMILES
82	smiles = Chem.MolToSmiles(mol)
83
84	# To MOL block
85	mol_block = Chem.MolToMolBlock(mol)
86
87	# To InChI
88	inchi = Chem.MolToInchi(mol)
89	```
90
91	Batch Processing:
92
93	For processing multiple molecules, use Supplier/Writer objects:
94
95	```python
96	# Read SDF files
97	suppl = Chem.SDMolSupplier('molecules.sdf')
98	for mol in suppl:
99	if mol is not None: # Check for parsing errors
100	# Process molecule
101	pass
102
103	# Read SMILES files
104	suppl = Chem.SmilesMolSupplier('molecules.smi', titleLine=False)
105
106	# For large files or compressed data
107	with gzip.open('molecules.sdf.gz') as f:
108	suppl = Chem.ForwardSDMolSupplier(f)
109	for mol in suppl:
110	# Process molecule
111	pass
112
113	# Multithreaded processing for large datasets
114	suppl = Chem.MultithreadedSDMolSupplier('molecules.sdf')
115
116	# Write molecules to SDF
117	writer = Chem.SDWriter('output.sdf')
118	for mol in molecules:
119	writer.write(mol)
120	writer.close()
121	```
122
123	Important Notes:
124	- All `MolFrom*` functions return `None` on failure with error messages
125	- Always check for `None` before processing molecules
126	- Molecules are automatically sanitized on import (validates valence, perceives aromaticity)
127
128	### 2. Molecular Sanitization and Validation
129
130	RDKit automatically sanitizes molecules during parsing, executing 13 steps including valence checking, aromaticity perception, and chirality assignment.
131
132	Sanitization Control:
133
134	```python
135	# Disable automatic

People who liked this also viewed

6 picks

Type

rdkit

244stars35forks

View on GitHub

How to install rdkit?

ai4protein/venusfactory2/rdkit

$npx -y skills add ai4protein/VenusFactory2 --skill rdkit

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# RDKit Cheminformatics Toolkit
2
3	## Overview
4
5	RDKit is a comprehensive cheminformatics library providing Python APIs for molecular analysis and manipulation. This skill provides guidance for reading/writing molecular structures, calculating descriptors, fingerprinting, substructure searching, chemical reactions, 2D/3D coordinate generation, and molecular visualization. Use this skill for drug discovery, computational chemistry, and cheminformatics research tasks.
6
7	## Quick Start and Project Scripts
8
9	Scripts live in `src/tools/bioinfo/rdkit/`. Each module is an executable atomic script with `if __name__ == "__main__"` (CLI via argparse).
10
11	\| Script \| Purpose \| Main \|
12	\|--------\|---------\|------\|
13	\| `molecular_properties.py` \| Properties: MW, LogP, TPSA, Lipinski, QED; single SMILES or batch file → CSV \| Yes (argparse) \|
14	\| `substructure_filter.py` \| Filter by SMARTS/SMILES include/exclude; pattern libs (functional-groups, rings, pains, privileged) \| Yes (argparse) \|
15	\| `similarity_search.py` \| Fingerprint similarity (Morgan, RDKit, MACCS, atompair, torsion); Tanimoto/Dice/cosine \| Yes (argparse) \|
16
17	Import from package:
18	```python
19	from src.tools.bioinfo.rdkit import (
20	calculate_properties,
21	process_single_molecule,
22	process_file,
23	filter_molecules,
24	create_pattern_query,
25	PATTERN_LIBRARIES,
26	generate_fingerprint,
27	similarity_search,
28	FINGERPRINT_METHODS,
29	)
30	# Example: single-molecule properties
31	props = calculate_properties(Chem.MolFromSmiles("CCO"))
32	# Example: substructure filter (mols from load_molecules_filter)
33	from src.tools.bioinfo.rdkit.substructure_filter import load_molecules as load_mols
34	mols = load_mols("molecules.smi")
35	filtered, info = filter_molecules(mols, include_patterns=[("benzene", Chem.MolFromSmarts("c1ccccc1"))])
36	# Example: similarity (query_mol + list of dicts with 'mol', 'smiles', 'name', 'index')
37	from src.tools.bioinfo.rdkit.similarity_search import load_molecules as load_db
38	db = load_db("database.smi")
39	hits = similarity_search(query_mol, db, method="morgan", threshold=0.7)
40	```
41
42	Run as CLI (each has `main()`):
43	```bash
44	python -m src.tools.bioinfo.rdkit.molecular_properties "CCO"
45	python -m src.tools.bioinfo.rdkit.molecular_properties --file molecules.smi --output props.csv
46	python -m src.tools.bioinfo.rdkit.substructure_filter molecules.smi --pattern "c1ccccc1" -o filtered.smi
47	python -m src.tools.bioinfo.rdkit.similarity_search "CCO" database.smi --threshold 0.7 -o hits.csv
48	```
49
50	References: `references/api_reference.md`, `references/descriptors_reference.md`, `references/smarts_patterns.md`.
51
52	## Core Capabilities
53
54	### 1. Molecular I/O and Creation
55
56	Reading Molecules:
57
58	Read molecular structures from various formats:
59
60	```python
61	from rdkit import Chem
62
63	# From SMILES strings
64	mol = Chem.MolFromSmiles('Cc1ccccc1') # Returns Mol object or None
65
66	# From MOL files
67	mol = Chem.MolFromMolFile('path/to/file.mol')
68
69	# From MOL blocks (string data)
70	mol = Chem.MolFromMolBlock(mol_block_string)
71
72	# From InChI
73	mol = Chem.MolFromInchi('InChI=1S/C6H6/c1-2-4-6-5-3-1/h1-6H')
74	```
75
76	Writing Molecules:
77
78	Convert molecules to text representations:
79
80	```python
81	# To canonical SMILES
82	smiles = Chem.MolToSmiles(mol)
83
84	# To MOL block
85	mol_block = Chem.MolToMolBlock(mol)
86
87	# To InChI
88	inchi = Chem.MolToInchi(mol)
89	```
90
91	Batch Processing:
92
93	For processing multiple molecules, use Supplier/Writer objects:
94
95	```python
96	# Read SDF files
97	suppl = Chem.SDMolSupplier('molecules.sdf')
98	for mol in suppl:
99	if mol is not None: # Check for parsing errors
100	# Process molecule
101	pass
102
103	# Read SMILES files
104	suppl = Chem.SmilesMolSupplier('molecules.smi', titleLine=False)
105
106	# For large files or compressed data
107	with gzip.open('molecules.sdf.gz') as f:
108	suppl = Chem.ForwardSDMolSupplier(f)
109	for mol in suppl:
110	# Process molecule
111	pass
112
113	# Multithreaded processing for large datasets
114	suppl = Chem.MultithreadedSDMolSupplier('molecules.sdf')
115
116	# Write molecules to SDF
117	writer = Chem.SDWriter('output.sdf')
118	for mol in molecules:
119	writer.write(mol)
120	writer.close()
121	```
122
123	Important Notes:
124	- All `MolFrom*` functions return `None` on failure with error messages
125	- Always check for `None` before processing molecules
126	- Molecules are automatically sanitized on import (validates valence, perceives aromaticity)
127
128	### 2. Molecular Sanitization and Validation
129
130	RDKit automatically sanitizes molecules during parsing, executing 13 steps including valence checking, aromaticity perception, and chirality assignment.
131
132	Sanitization Control:
133
134	```python
135	# Disable automatic

People who liked this also viewed

6 picks

Type