liveBook · Manning

Warm start

As explained in Build AI Drug Discovery Pipelines

A warm-start evaluation setting allows test entities to have appeared in training while withholding the specific test combinations.

Warm-start evaluation tests combinations whose compounds and proteins may have appeared during training, while withholding the specific compound–protein pairs. It measures whether a model can predict new relationships among known entities. section 11.5.5

This setting reflects practical drug-discovery tasks such as predicting known drugs against new targets or new drugs against known targets. Because repeated entity exposure can support memorization, warm-start scores may overstate performance on genuinely new chemistry or targets. section 11.5.5

Compare warm-start evaluation with stricter alternatives:

  • Cold-drug splits hold out entire drugs; cold-target splits hold out proteins; cold-drug/cold-target splits hold out both partners. section 11.5.5

How Build AI Drug Discovery Pipelines covers it

How deep it goes
Defines the split and its rationale, then contrasts it with cold-start protocols without detailing implementation steps.
Examples
Uses Davis and KIBA affinity datasets, including known drugs tested against new targets and new drugs against known targets.
What you can do after
Evaluate affinity predictions while recognizing that warm-start scores may not represent performance on novel chemistry or targets.
How it connects
Connects drug-target affinity prediction with cold-start evaluation, memorization, and generalization to unseen entities.

This explainer was generated by Manning's knowledge graph from the text of Build AI Drug Discovery Pipelines. The section links point to the parts of the book it draws on.