datamol

Name: datamol
Author: LeonChaoX

698stars61forks

How to install datamol?

leonchaox/qinyan-academic-skills/datamol

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

Installs into the current project.

Summary

Pythonic wrapper around RDKit with simplified interface and sensible defaults. Preferred for standard drug discovery including SMILES parsing, standardization, descriptors, fingerprints, clustering, 3D conformers, parallel processing. Returns native rdkit.Chem.Mol objects. For ad

Files · 1

View on GitHub

SKILL.md

1	# Datamol Cheminformatics Skill
2
3	## Overview
4
5	Datamol is a Python library that provides a lightweight, Pythonic abstraction layer over RDKit for molecular cheminformatics. Simplify complex molecular operations with sensible defaults, efficient parallelization, and modern I/O capabilities. All molecular objects are native `rdkit.Chem.Mol` instances, ensuring full compatibility with the RDKit ecosystem.
6
7	Key capabilities:
8	- Molecular format conversion (SMILES, SELFIES, InChI)
9	- Structure standardization and sanitization
10	- Molecular descriptors and fingerprints
11	- 3D conformer generation and analysis
12	- Clustering and diversity selection
13	- Scaffold and fragment analysis
14	- Chemical reaction application
15	- Visualization and alignment
16	- Batch processing with parallelization
17	- Cloud storage support via fsspec
18
19	## Installation and Setup
20
21	Guide users to install datamol:
22
23	```bash
24	uv pip install datamol
25	```
26
27	Import convention:
28	```python
29	import datamol as dm
30	```
31
32	## Core Workflows
33
34	### 1. Basic Molecule Handling
35
36	Creating molecules from SMILES:
37	```python
38	import datamol as dm
39
40	# Single molecule
41	mol = dm.to_mol("CCO") # Ethanol
42
43	# From list of SMILES
44	smiles_list = ["CCO", "c1ccccc1", "CC(=O)O"]
45	mols = [dm.to_mol(smi) for smi in smiles_list]
46
47	# Error handling
48	mol = dm.to_mol("invalid_smiles") # Returns None
49	if mol is None:
50	print("Failed to parse SMILES")
51	```
52
53	Converting molecules to SMILES:
54	```python
55	# Canonical SMILES
56	smiles = dm.to_smiles(mol)
57
58	# Isomeric SMILES (includes stereochemistry)
59	smiles = dm.to_smiles(mol, isomeric=True)
60
61	# Other formats
62	inchi = dm.to_inchi(mol)
63	inchikey = dm.to_inchikey(mol)
64	selfies = dm.to_selfies(mol)
65	```
66
67	Standardization and sanitization (always recommend for user-provided molecules):
68	```python
69	# Sanitize molecule
70	mol = dm.sanitize_mol(mol)
71
72	# Full standardization (recommended for datasets)
73	mol = dm.standardize_mol(
74	mol,
75	disconnect_metals=True,
76	normalize=True,
77	reionize=True
78	)
79
80	# For SMILES strings directly
81	clean_smiles = dm.standardize_smiles(smiles)
82	```
83
84	### 2. Reading and Writing Molecular Files
85
86	Refer to `references/io_module.md` for comprehensive I/O documentation.
87
88	Reading files:
89	```python
90	# SDF files (most common in chemistry)
91	df = dm.read_sdf("compounds.sdf", mol_column='mol')
92
93	# SMILES files
94	df = dm.read_smi("molecules.smi", smiles_column='smiles', mol_column='mol')
95
96	# CSV with SMILES column
97	df = dm.read_csv("data.csv", smiles_column="SMILES", mol_column="mol")
98
99	# Excel files
100	df = dm.read_excel("compounds.xlsx", sheet_name=0, mol_column="mol")
101
102	# Universal reader (auto-detects format)
103	df = dm.open_df("file.sdf") # Works with .sdf, .csv, .xlsx, .parquet, .json
104	```
105
106	Writing files:
107	```python
108	# Save as SDF
109	dm.to_sdf(mols, "output.sdf")
110	# Or from DataFrame
111	dm.to_sdf(df, "output.sdf", mol_column="mol")
112
113	# Save as SMILES file
114	dm.to_smi(mols, "output.smi")
115
116	# Excel with rendered molecule images
117	dm.to_xlsx(df, "output.xlsx", mol_columns=["mol"])
118	```
119
120	Remote file support (S3, GCS, HTTP):
121	```python
122	# Read from cloud storage
123	df = dm.read_sdf("s3://bucket/compounds.sdf")
124	df = dm.read_csv("https://example.com/data.csv")
125
126	# Write to cloud storage
127	dm.to_sdf(mols, "s3://bucket/output.sdf")
128	```
129
130	### 3. Molecular Descriptors and Properties
131
132	Refer to `references/descriptors_viz.md` for detailed descriptor documentation.
133
134	Computing descriptors for a single molecule:
135	```python
136	# Get standard descriptor set
137	descriptors = dm.descriptors.compute_many_descriptors(mol)
138	# Returns: {'mw': 46.07, 'logp': -0.03, 'hbd': 1, 'hba': 1,
139	# 'tpsa': 20.23, 'n_aromatic_atoms': 0, ...}
140	```
141
142	Batch descriptor computation (recommended for datasets):
143	```python
144	# Compute for all molecules in parallel
145	desc_df = dm.descriptors.batch_compute_many_descriptors(
146	mols,
147	n_jobs=-1, # Use all CPU cores
148	progress=True # Show progress bar
149	)
150	```
151
152	Specific descriptors:
153	```python
154	# Aromaticity
155	n_aromatic = dm.descriptors.n_aromatic_atoms(mol)
156	aromatic_ratio = dm.descriptors.n_aromatic_atoms_proportion(mol)
157
158	# Stereochemistry
159	n_stereo = dm.descriptors.n_stereo_centers(mol)
160	n_unspec = dm.descriptors.n_stereo_centers_unspecified(mol)
161
162	# Flexibility
163	n_rigid = dm.descriptors.n_rigid_bonds(mol)
164	```
165
166	Drug-likeness filtering (Lipinski's Rule of Five):
167	```python
168	# Filter compounds
169	def is_druglike(mol):
170	desc = dm.descriptors.compute_many_descriptors(mol)
171	return (
172	desc['mw'] <= 500 and
173	desc['logp'] <= 5 and
174	desc['hbd'] <= 5 and
175	desc['hba'] <= 10
176	)
177
178	druglike_mols = [mol for mol in mo

People who liked this also viewed

6 picks

Type

datamol

698stars61forks

View on GitHub

How to install datamol?

leonchaox/qinyan-academic-skills/datamol

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

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Datamol Cheminformatics Skill
2
3	## Overview
4
5	Datamol is a Python library that provides a lightweight, Pythonic abstraction layer over RDKit for molecular cheminformatics. Simplify complex molecular operations with sensible defaults, efficient parallelization, and modern I/O capabilities. All molecular objects are native `rdkit.Chem.Mol` instances, ensuring full compatibility with the RDKit ecosystem.
6
7	Key capabilities:
8	- Molecular format conversion (SMILES, SELFIES, InChI)
9	- Structure standardization and sanitization
10	- Molecular descriptors and fingerprints
11	- 3D conformer generation and analysis
12	- Clustering and diversity selection
13	- Scaffold and fragment analysis
14	- Chemical reaction application
15	- Visualization and alignment
16	- Batch processing with parallelization
17	- Cloud storage support via fsspec
18
19	## Installation and Setup
20
21	Guide users to install datamol:
22
23	```bash
24	uv pip install datamol
25	```
26
27	Import convention:
28	```python
29	import datamol as dm
30	```
31
32	## Core Workflows
33
34	### 1. Basic Molecule Handling
35
36	Creating molecules from SMILES:
37	```python
38	import datamol as dm
39
40	# Single molecule
41	mol = dm.to_mol("CCO") # Ethanol
42
43	# From list of SMILES
44	smiles_list = ["CCO", "c1ccccc1", "CC(=O)O"]
45	mols = [dm.to_mol(smi) for smi in smiles_list]
46
47	# Error handling
48	mol = dm.to_mol("invalid_smiles") # Returns None
49	if mol is None:
50	print("Failed to parse SMILES")
51	```
52
53	Converting molecules to SMILES:
54	```python
55	# Canonical SMILES
56	smiles = dm.to_smiles(mol)
57
58	# Isomeric SMILES (includes stereochemistry)
59	smiles = dm.to_smiles(mol, isomeric=True)
60
61	# Other formats
62	inchi = dm.to_inchi(mol)
63	inchikey = dm.to_inchikey(mol)
64	selfies = dm.to_selfies(mol)
65	```
66
67	Standardization and sanitization (always recommend for user-provided molecules):
68	```python
69	# Sanitize molecule
70	mol = dm.sanitize_mol(mol)
71
72	# Full standardization (recommended for datasets)
73	mol = dm.standardize_mol(
74	mol,
75	disconnect_metals=True,
76	normalize=True,
77	reionize=True
78	)
79
80	# For SMILES strings directly
81	clean_smiles = dm.standardize_smiles(smiles)
82	```
83
84	### 2. Reading and Writing Molecular Files
85
86	Refer to `references/io_module.md` for comprehensive I/O documentation.
87
88	Reading files:
89	```python
90	# SDF files (most common in chemistry)
91	df = dm.read_sdf("compounds.sdf", mol_column='mol')
92
93	# SMILES files
94	df = dm.read_smi("molecules.smi", smiles_column='smiles', mol_column='mol')
95
96	# CSV with SMILES column
97	df = dm.read_csv("data.csv", smiles_column="SMILES", mol_column="mol")
98
99	# Excel files
100	df = dm.read_excel("compounds.xlsx", sheet_name=0, mol_column="mol")
101
102	# Universal reader (auto-detects format)
103	df = dm.open_df("file.sdf") # Works with .sdf, .csv, .xlsx, .parquet, .json
104	```
105
106	Writing files:
107	```python
108	# Save as SDF
109	dm.to_sdf(mols, "output.sdf")
110	# Or from DataFrame
111	dm.to_sdf(df, "output.sdf", mol_column="mol")
112
113	# Save as SMILES file
114	dm.to_smi(mols, "output.smi")
115
116	# Excel with rendered molecule images
117	dm.to_xlsx(df, "output.xlsx", mol_columns=["mol"])
118	```
119
120	Remote file support (S3, GCS, HTTP):
121	```python
122	# Read from cloud storage
123	df = dm.read_sdf("s3://bucket/compounds.sdf")
124	df = dm.read_csv("https://example.com/data.csv")
125
126	# Write to cloud storage
127	dm.to_sdf(mols, "s3://bucket/output.sdf")
128	```
129
130	### 3. Molecular Descriptors and Properties
131
132	Refer to `references/descriptors_viz.md` for detailed descriptor documentation.
133
134	Computing descriptors for a single molecule:
135	```python
136	# Get standard descriptor set
137	descriptors = dm.descriptors.compute_many_descriptors(mol)
138	# Returns: {'mw': 46.07, 'logp': -0.03, 'hbd': 1, 'hba': 1,
139	# 'tpsa': 20.23, 'n_aromatic_atoms': 0, ...}
140	```
141
142	Batch descriptor computation (recommended for datasets):
143	```python
144	# Compute for all molecules in parallel
145	desc_df = dm.descriptors.batch_compute_many_descriptors(
146	mols,
147	n_jobs=-1, # Use all CPU cores
148	progress=True # Show progress bar
149	)
150	```
151
152	Specific descriptors:
153	```python
154	# Aromaticity
155	n_aromatic = dm.descriptors.n_aromatic_atoms(mol)
156	aromatic_ratio = dm.descriptors.n_aromatic_atoms_proportion(mol)
157
158	# Stereochemistry
159	n_stereo = dm.descriptors.n_stereo_centers(mol)
160	n_unspec = dm.descriptors.n_stereo_centers_unspecified(mol)
161
162	# Flexibility
163	n_rigid = dm.descriptors.n_rigid_bonds(mol)
164	```
165
166	Drug-likeness filtering (Lipinski's Rule of Five):
167	```python
168	# Filter compounds
169	def is_druglike(mol):
170	desc = dm.descriptors.compute_many_descriptors(mol)
171	return (
172	desc['mw'] <= 500 and
173	desc['logp'] <= 5 and
174	desc['hbd'] <= 5 and
175	desc['hba'] <= 10
176	)
177
178	druglike_mols = [mol for mol in mo

People who liked this also viewed

6 picks

Type