How does the skill handle data splitting?

It provides scientifically meaningful splits such as scaffold-based, cold-drug, and cold-target splits to ensure realistic model evaluation.

PyTDC (Therapeutics Data Commons) is an open-science platform providing datasets and benchmarks for machine learning in drug discovery and development.

Can I use PyTDC for molecular optimization?

Yes, it includes 'oracles' which are scoring functions used to evaluate and optimize the properties of generated molecules.

Which therapeutic tasks does this skill support?

It supports a wide range of tasks including ADME prediction, toxicity forecasting, drug-target interactions, and molecular generation.

Is PyTDC suitable for clinical trial prediction?

Yes, PyTDC includes datasets for clinical trial outcomes and toxicity (ClinTox) to assist in late-stage development modeling.

Therapeutics Data Commons (PyTDC)

Name: Therapeutics Data Commons (PyTDC)
Author: pur3v4d3r

bypur3v4d3r

•

数据科学与机器学习

Facilitates drug discovery and therapeutic machine learning by providing AI-ready datasets, benchmarks, and molecular evaluation oracles.

PyTDC is a comprehensive open-science platform designed to streamline the application of machine learning in therapeutics. It offers a standardized interface for accessing curated datasets spanning the entire drug discovery pipeline—including ADME, toxicity, and drug-target interactions—along with specialized data splits like scaffold and cold-start. By integrating molecular oracles for property optimization and evaluation metrics for benchmarking, PyTDC empowers developers to build, test, and compare AI models for pharmacological prediction and novel molecule generation with scientific rigor.

主要功能

011 GitHub stars

02AI-ready datasets for ADME, toxicity, and bioactivity prediction

03Built-in molecular format conversion and chemical data processing utilities

04Specialized data splits including scaffold, cold-drug, and temporal methods

05Standardized benchmark groups for systematic model evaluation

06Molecular generation oracles for scoring drug-likeness and binding affinity

使用场景

01Benchmarking new machine learning architectures on standardized pharmaceutical datasets

02Generating and optimizing novel molecules with specific desired chemical properties

03Predicting pharmacokinetic properties (ADMET) for candidate drug molecules

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add pur3v4d3r/pur3-pkb-codebase pytdc

For use in Claude.ai and ChatGPT

Download Skill