skills.rest
DrugClaw
DrugClawOFFICIAL
Explore all skills by this author →
WebsiteGitHub

pharma-ml-tools

Profile compound libraries and featurize molecules for pharmaceutical machine-learning workflows.

126|8|Updated Mar 10, 2026
Search Tags:#machine learning#drug discovery#medicinal chemistry#cheminformatics#dataset preparation#molecular featurization
One-click install
npx skills add https://github.com/DrugClaw/DrugClaw --skill pharma-ml-tools
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: pharma-ml-tools
Source: https://github.com/DrugClaw/DrugClaw/tree/main/skills/pharma/pharma-ml-tools
Command: npx skills add https://github.com/DrugClaw/DrugClaw --skill pharma-ml-tools

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill streamlines the process of preparing and featurizing molecular data for machine learning in drug discovery, reducing manual effort and ensuring reproducibility.

Core Features & Use Cases

  • Library Profiling: Standardize and analyze compound libraries for duplicates, invalid structures, and scaffold diversity using datamol.
  • Molecular Featurization: Generate machine-learning-ready features (e.g., ECFP, MACCS, RDKit2D) using molfeat.
  • Benchmark Dataset Fetching: Download and prepare standardized datasets for ADME, toxicity, DTI, and DDI studies using PyTDC.
  • Medicinal Chemistry Screening: Apply common rules and alert filters (e.g., Rule of Five, lead-likeness) using medchem.
  • Use Case: Prepare a library of drug candidates for QSAR modeling by profiling, featurizing, and screening them for potential liabilities.

Quick Start

Use the pharma-ml-tools skill to profile the compound library located at 'libraries/kinase_hits.csv' using the 'smiles' column and save the output to 'pharma_ml/kinase_hits_profile.csv'.

Frequently Asked Questions about pharma-ml-tools

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I prepare a compound library for QSAR modeling?▼

Prepare a compound library for QSAR modeling by profiling structures for duplicates, featurizing molecules using ECFP or MACCS, and applying medicinal chemistry filters to generate an ML-ready dataset.

What is molecular featurization for drug discovery datasets?▼

Molecular featurization for drug discovery datasets converts chemical structures into machine-learning-ready numerical features like ECFP, MACCS, and RDKit2D representations to enable predictive model training.

How do I download benchmark datasets for ADME and toxicity prediction?▼

Download benchmark datasets for ADME and toxicity prediction using the PyTDC library to fetch standardized data for ADME, toxicity, DTI, and DDI studies, ensuring reproducible model evaluation.

Can I filter compound libraries using Rule of Five and lead-likeness rules?▼

Filter compound libraries using Rule of Five and lead-likeness rules by applying medicinal chemistry alert functions to identify and remove structures with potential liabilities before ML modeling.

Do I need Python and specific cheminformatics packages for molecular dataset preparation?▼

Molecular dataset preparation requires Python 3 with pandas, numpy, datamol, molfeat, medchem, and PyTDC installed to enable library profiling, featurization, and benchmark dataset acquisition.

What's the best way to analyze scaffold diversity in a compound library?▼

Analyze scaffold diversity in a compound library by using datamol to profile structural uniqueness, detect invalid structures, and evaluate the distribution of chemical scaffolds for ML triage.

Related Skills

Explore More Education & Research Skills→
jasrajtulsijasrajtulsi

medchem

Filter compound libraries with medicinal chemistry rules and structural alerts.

Community
Advanced
HolobiomicsLabHolobiomicsLab
4

deepchem

Load molecular data, featurize it, and train predictive models.

Official
Advanced
RamanEbrahimiRamanEbrahimi
3

molfeat

Convert SMILES strings into numerical features for machine learning.

Community
Advanced
InternScienceInternScience
28

molclaw-compound-retrieve

Retrieve canonical SMILES strings from PubChem for batch compound names.

Official
Basic