Load TDC drug discovery datasets with standardized splits and evaluators.
npx skills add https://github.com/jimmc414/Kosmos --skill pytdc-jimmc414
Please help me install this Agent Skill. Skill: pytdc Source: https://github.com/jimmc414/Kosmos/tree/main/kosmos-claude-scientific-skills/scientific-skills/pytdc Command: npx skills add https://github.com/jimmc414/Kosmos --skill pytdc-jimmc414
This Skill provides access to a comprehensive suite of curated datasets and standardized benchmarks for drug discovery and development, streamlining machine learning model development and evaluation.
Use the pytdc skill to load the Caco2_Wang ADME dataset and get a scaffold split.
High-intent search queries and answers about installing and using this skill.
You can load drug discovery datasets for machine learning benchmarks by accessing the Therapeutics Data Commons through this Skill, which provides curated single-instance and multi-instance prediction tasks with standardized data splits.
Yes, this Skill provides standardized data splitting strategies for ADME and toxicity prediction, including scaffold splits for molecular generalization and cold-splits for evaluating novel drug-target interactions.
Yes, you can access drug-target interaction datasets like BindingDB with a cold-drug split to ensure your machine learning model generalizes to novel compounds during therapeutic development.
Yes, you need the PyTDC library installed to access and manipulate the cheminformatics dataset benchmarks, as this Skill relies on it to load Therapeutics Data Commons datasets and evaluators.
The Therapeutics Data Commons supports single-instance prediction tasks like ADME and toxicity, multi-instance prediction tasks like drug-target and drug-drug interactions, and molecular generation tasks.
You evaluate molecular generation models using built-in evaluators from the Therapeutics Data Commons, which provide standardized metrics for systematic comparison across pre-defined benchmark groups.
DrugClawProfile compound libraries and featurize molecules for pharmaceutical machine-learning workflows.
qinyan-aiLoad PyTDC datasets and benchmarks for drug discovery experiments.
SciMate-AIPredict molecular properties and materials descriptors from chemical data using DeepChem.
pur3v4d3rLoad PyTDC drug discovery datasets and generate scaffold, random, or cold splits.
