Who it is forA medicinal chemist choosing which analogs of a lead compound to synthesise next.
The problemSynthesising and assaying every analog is slow and expensive.
The outcomePredicted log10(IC50) values for each analog, so you can prioritise which ones to make.
What you need first
- Python in a fresh environment and pip
- An NVIDIA GPU for the [cuda] install; CPU-only works but is significantly slower
- Input YAML files describing the biomolecules (format in docs/prediction.md)
- Network access for the optional MSA server (--use_msa_server)
Steps
- Install from GitHub for daily updates if you want the newest code:
git clone https://github.com/jwohlwend/boltz.gitthencd boltz; pip install -e .[cuda] - Write YAML inputs for the lead and each modified analog against the same target, requesting affinity
- Run
boltz predict input_path --use_msa_serveron the directory of inputs - Compare affinity_pred_value, reported as log10(IC50) derived from an IC50 in micromolar, across analogs
- Use the results to rank which analogs to synthesise first
Prompt for your agent
Using jwohlwend/boltz, predict affinity for my lead compound <SMILES> and these analogs <list> against <target>, then make a table of affinity_pred_value per analog with a note on which modifications helped.
Caveat. Updated Boltz-2 evaluation and training code are marked as coming soon in the README.
Use this in your AI agent
Ask your agent
Using the Deduped MCP, get the setup steps for jwohlwend/boltz and "Compare small changes to a lead molecule"
Add Deduped to Claude Code once
claude mcp add --transport http deduped https://dedupedai.com/mcp
Every Monday: the 5 hottest repos and exactly how to use them. Free, one email a week, unsubscribe anytime.
Get them freeShare: X LinkedIn Hacker News Email Copy link
Third-party repos we don't own. AI-written use cases can be wrong. Disclaimer