Load Therapeutics Data Commons datasets for drug discovery machine learning tasks.
npx skills add https://github.com/hxk622/TokenDance --skill pytdc-hxk622
Please help me install this Agent Skill. Skill: pytdc Source: https://github.com/hxk622/TokenDance/tree/main/backend/app/skills/builtin/scientific/chemistry/pytdc Command: npx skills add https://github.com/hxk622/TokenDance --skill pytdc-hxk622
This Skill provides access to a comprehensive collection of curated datasets and benchmarks for drug discovery and development, streamlining machine learning model development and evaluation.
Caco2_Wang dataset and training a predictive model.Use the pytdc skill to load the 'Caco2_Wang' ADME dataset and get its scaffold split.
High-intent search queries and answers about installing and using this skill.
To load ADME and toxicity datasets for drug discovery, you can access the Therapeutics Data Commons and retrieve specific collections like the Caco2_Wang dataset. This provides curated data ready for training machine learning models.
Standardized data splits for pharmacology benchmarks include scaffold splits and cold-split strategies. These methods ensure systematic model evaluation by structuring training and test sets according to molecular frameworks or entity exclusions.
Yes, Therapeutics Data Commons supports multi-instance prediction tasks including drug-target interactions (DTI) and drug-drug interactions (DDI). It provides structured datasets to evaluate machine learning models on complex pharmacological relationships.
You can evaluate molecular generation models for drug discovery using standardized benchmarks and molecular oracles provided by the Therapeutics Data Commons. These tools assess generated molecules against specific pharmacological properties.
Yes, the pytdc skill provides pre-defined benchmark groups for systematic model comparison in drug discovery. These benchmarks offer standardized datasets, splits, and evaluation metrics to consistently assess machine learning performance.
The Therapeutics Data Commons supports single-instance prediction for properties like ADME and toxicity, multi-instance prediction for drug-target interactions, and molecular generation tasks for machine learning in pharmacology.
mims-harvardRetrieve and validate chemical compound data from PubChem and ChEMBL APIs.
KaiserWhoLearnsTrain and evaluate machine learning models across scikit-learn, PyTorch, and TensorFlow.
pradyumnasagarCalculate molecular fingerprints and physicochemical descriptors from SMILES using RDKit.
timurgaleevBenchmark AI models across Claude, GPT, and Gemini for latency, cost, and quality.