rdkit-cli 0.2.0

A comprehensive CLI tool for RDKit cheminformatics operations

rdkit-cli

A comprehensive, high-performance CLI tool wrapping RDKit functionality for cheminformatics workflows.

Features

• 19 Command Categories: descriptors, fingerprints, filter, convert, standardize, similarity, conformers, reactions, scaffold, enumerate, fragment, diversity, mcs, depict, stats, split, sample, deduplicate, validate
• Multiple Input/Output Formats: CSV, TSV, SMI, SDF, Parquet
• Parallel Processing: Efficient multi-core support via ProcessPoolExecutor
• Ninja-style Progress: Real-time progress display with speed and ETA

Installation

pip install rdkit-cli

Or with uv:

uv add rdkit-cli

Quick Start

# Compute molecular descriptors
rdkit-cli descriptors compute -i molecules.csv -o descriptors.csv -d MolWt,MolLogP,TPSA

# Generate fingerprints
rdkit-cli fingerprints compute -i molecules.csv -o fingerprints.csv --type morgan

# Filter by drug-likeness
rdkit-cli filter druglike -i molecules.csv -o filtered.csv --rule lipinski

# Standardize molecules
rdkit-cli standardize -i molecules.csv -o standardized.csv --cleanup --neutralize

# Similarity search
rdkit-cli similarity search -i library.csv -o hits.csv --query "c1ccccc1" --threshold 0.7

Commands

descriptors

Compute molecular descriptors.

# List available descriptors
rdkit-cli descriptors list
rdkit-cli descriptors list --all

# Compute specific descriptors
rdkit-cli descriptors compute -i input.csv -o output.csv -d MolWt,MolLogP,TPSA

# Compute all descriptors
rdkit-cli descriptors compute -i input.csv -o output.csv --all

fingerprints

Generate molecular fingerprints.

# List available fingerprint types
rdkit-cli fingerprints list

# Compute Morgan fingerprints (default)
rdkit-cli fingerprints compute -i input.csv -o output.csv --type morgan

# With options
rdkit-cli fingerprints compute -i input.csv -o output.csv \
    --type morgan --radius 3 --bits 4096 --use-chirality

Supported types: morgan, maccs, rdkit, atompair, torsion, pattern

filter

Filter molecules by various criteria.

# Substructure filter
rdkit-cli filter substructure -i input.csv -o output.csv --smarts "c1ccccc1"
rdkit-cli filter substructure -i input.csv -o output.csv --smarts "c1ccccc1" --exclude

# Property filter
rdkit-cli filter property -i input.csv -o output.csv --rule "MolWt < 500"

# Drug-likeness filters
rdkit-cli filter druglike -i input.csv -o output.csv --rule lipinski
rdkit-cli filter druglike -i input.csv -o output.csv --rule veber
rdkit-cli filter druglike -i input.csv -o output.csv --rule ghose

# PAINS filter
rdkit-cli filter pains -i input.csv -o output.csv

convert

Convert between molecular file formats.

# Auto-detect formats from extensions
rdkit-cli convert -i molecules.csv -o molecules.sdf

# Explicit format specification
rdkit-cli convert -i molecules.csv -o molecules.smi --out-format smi

Supported formats: csv, tsv, smi, sdf, parquet

standardize

Standardize and canonicalize molecules.

# Basic standardization
rdkit-cli standardize -i input.csv -o output.csv

# With options
rdkit-cli standardize -i input.csv -o output.csv \
    --cleanup --neutralize --fragment-parent

similarity

Compute molecular similarity.

# Similarity search
rdkit-cli similarity search -i library.csv -o hits.csv \
    --query "CCO" --threshold 0.7

# Similarity matrix
rdkit-cli similarity matrix -i molecules.csv -o matrix.csv \
    --metric tanimoto

# Clustering
rdkit-cli similarity cluster -i molecules.csv -o clustered.csv \
    --cutoff 0.5

conformers

Generate and optimize 3D conformers.

# Generate conformers
rdkit-cli conformers generate -i input.csv -o output.sdf --num 10

# Optimize conformers
rdkit-cli conformers optimize -i input.sdf -o optimized.sdf --force-field mmff

reactions

Apply chemical reactions and transformations.

# SMIRKS transformation
rdkit-cli reactions transform -i input.csv -o output.csv \
    --smirks "[OH:1]>>[O-:1]"

# Reaction enumeration
rdkit-cli reactions enumerate -i reactants.csv -o products.csv \
    --template "reaction.rxn"

scaffold

Extract molecular scaffolds.

# Murcko scaffolds
rdkit-cli scaffold murcko -i input.csv -o scaffolds.csv

# Generic scaffolds
rdkit-cli scaffold murcko -i input.csv -o scaffolds.csv --generic

# Scaffold decomposition
rdkit-cli scaffold decompose -i input.csv -o decomposed.csv

enumerate

Enumerate molecular variants.

# Stereoisomers
rdkit-cli enumerate stereoisomers -i input.csv -o isomers.csv --max-isomers 32

# Tautomers
rdkit-cli enumerate tautomers -i input.csv -o tautomers.csv --max-tautomers 50

# Canonical tautomer
rdkit-cli enumerate canonical-tautomer -i input.csv -o canonical.csv

fragment

Fragment molecules.

# BRICS fragmentation
rdkit-cli fragment brics -i input.csv -o fragments.csv

# RECAP fragmentation
rdkit-cli fragment recap -i input.csv -o fragments.csv

# Functional group extraction
rdkit-cli fragment functional-groups -i input.csv -o groups.csv

# Fragment frequency analysis
rdkit-cli fragment analyze -i fragments.csv -o analysis.csv

diversity

Analyze and select diverse molecules.

# Pick diverse subset
rdkit-cli diversity pick -i input.csv -o diverse.csv -k 100

# Analyze diversity
rdkit-cli diversity analyze -i input.csv

mcs

Find Maximum Common Substructure.

# Find MCS across molecules
rdkit-cli mcs -i molecules.csv -o mcs_result.csv

# With options
rdkit-cli mcs -i molecules.csv -o mcs_result.csv \
    --timeout 60 --atom-compare elements

depict

Generate molecular depictions.

# Single molecule
rdkit-cli depict single --smiles "c1ccccc1" -o benzene.svg

# Batch depiction
rdkit-cli depict batch -i molecules.csv -o images/ -f svg

# Grid image
rdkit-cli depict grid -i molecules.csv -o grid.svg --mols-per-row 4

stats

Calculate dataset statistics.

# Basic statistics
rdkit-cli stats -i molecules.csv -o stats.json --format json

# Specific properties
rdkit-cli stats -i molecules.csv -p MolWt,LogP,TPSA

# List available properties
rdkit-cli stats -i molecules.csv --list-properties

split

Split files into smaller chunks.

# Split into N files
rdkit-cli split -i large.csv -o chunks/ -c 10

# Split by chunk size
rdkit-cli split -i large.csv -o chunks/ -s 1000

# With custom prefix
rdkit-cli split -i large.csv -o chunks/ -c 5 --prefix molecules

sample

Randomly sample molecules.

# Sample by count
rdkit-cli sample -i molecules.csv -o sample.csv -k 100 --seed 42

# Sample by fraction
rdkit-cli sample -i molecules.csv -o sample.csv -f 0.1

# Memory-efficient streaming (reservoir sampling)
rdkit-cli sample -i huge.csv -o sample.csv -k 1000 --stream

deduplicate

Remove duplicate molecules.

# Deduplicate by canonical SMILES (default)
rdkit-cli deduplicate -i molecules.csv -o unique.csv

# Deduplicate by InChIKey
rdkit-cli deduplicate -i molecules.csv -o unique.csv -b inchikey

# Deduplicate by scaffold
rdkit-cli deduplicate -i molecules.csv -o unique.csv -b scaffold

# Keep last occurrence instead of first
rdkit-cli deduplicate -i molecules.csv -o unique.csv --keep last

validate

Validate molecular structures.

# Basic validation
rdkit-cli validate -i molecules.csv -o validated.csv

# Output only valid molecules
rdkit-cli validate -i molecules.csv -o valid.csv --valid-only

# With constraints
rdkit-cli validate -i molecules.csv -o validated.csv \
    --max-atoms 100 --max-rings 8

# Check allowed elements
rdkit-cli validate -i molecules.csv -o validated.csv \
    --allowed-elements C,H,N,O,S,F,Cl

# Check stereo and show summary
rdkit-cli validate -i molecules.csv -o validated.csv \
    --check-stereo --summary

Global Options

Option	Description
-n, --ncpu N	Number of CPUs (-1 = all, default: -1)
-i, --input FILE	Input file
-o, --output FILE	Output file
--smiles-column COL	SMILES column name (default: "smiles")
--name-column COL	Name column (optional)
--no-header	Input has no header row
-q, --quiet	Suppress progress output
-V, --version	Show version
-h, --help	Show help

Input/Output Formats

Format	Extension	Notes
CSV	.csv	Comma-separated, with header
TSV	.tsv	Tab-separated, with header
SMI	.smi	SMILES format, space-separated
SDF	.sdf	Structure-Data File
Parquet	.parquet	Apache Parquet format

Examples

Cheminformatics Pipeline

# 1. Validate and filter input
rdkit-cli validate -i raw.csv -o validated.csv --valid-only

# 2. Deduplicate
rdkit-cli deduplicate -i validated.csv -o unique.csv -b inchikey

# 3. Standardize molecules
rdkit-cli standardize -i unique.csv -o std.csv --cleanup --neutralize

# 4. Filter by drug-likeness
rdkit-cli filter druglike -i std.csv -o druglike.csv --rule lipinski

# 5. Compute descriptors
rdkit-cli descriptors compute -i druglike.csv -o desc.csv -d MolWt,MolLogP,TPSA,HBD,HBA

# 6. Get dataset statistics
rdkit-cli stats -i druglike.csv -o stats.json --format json

# 7. Select diverse subset
rdkit-cli diversity pick -i druglike.csv -o diverse.csv -k 500

# 8. Generate depictions
rdkit-cli depict grid -i diverse.csv -o library.svg --mols-per-row 10

Similarity Screening

# Search for similar compounds
rdkit-cli similarity search -i library.csv -o hits.csv \
    --query "CC(=O)Oc1ccccc1C(=O)O" \
    --threshold 0.6 \
    --type morgan

# Cluster results
rdkit-cli similarity cluster -i hits.csv -o clustered.csv --cutoff 0.4

Scaffold Analysis

# Extract scaffolds
rdkit-cli scaffold murcko -i library.csv -o scaffolds.csv

# Analyze scaffold diversity
rdkit-cli diversity analyze -i scaffolds.csv --smiles-column scaffold

Large Dataset Processing

# Sample from a huge dataset
rdkit-cli sample -i huge_library.csv -o sample.csv -k 10000 --stream

# Split for parallel processing
rdkit-cli split -i library.csv -o batches/ -c 10

# Process batches in parallel (using xargs)
ls batches/*.csv | xargs -P 4 -I {} rdkit-cli descriptors compute -i {} -o {}.desc.csv -d MolWt,LogP

Development

# Clone repository
git clone https://github.com/vitruves/rdkit-cli
cd rdkit-cli

# Install with dev dependencies
uv sync --dev

# Run tests
uv run pytest

# Run with coverage
uv run pytest --cov=rdkit_cli

License

Apache 2.0