Challenges in predicting cell permeability of PROTAC molecules
The Challenges of PROTAC Permeability Prediction
Digital Libraries
Summary
Predicting how well certain molecules called PROTACs can enter cells is tough because there isn’t enough clear data. The authors used a new method combining AI and expert review to gather more data from scientific papers, greatly increasing the number of examples available. They found that existing prediction models work well only on data similar to what they were trained on, but not on new types of PROTACs. This means more and better data sharing in science is needed to improve these prediction tools.
What this means in practice
- •For drug development teams: Improve selection of PROTAC candidates by using better-trained models on expanded permeability data sets.
- •For chemical data curators: Adopt expert-in-the-loop workflows to extract reliable chemical measurement data from published literature for database enhancement.
Authors
Lukáš Hudec, Rocío Mercado
Abstract
Cell permeability is a key bottleneck for PROTAC development, and public data available to model it is scarce and inconsistent. We adapt an expert-in-the-loop LLM extraction workflow to mine PAMPA measurements from the primary literature, recovering image-only structures by optical chemical structure recognition and hand-verifying every record, expanding the public record from 31 PROTACs to 87. Ridge models trained on PROTAC-DB 3.0 reach $R^2 = 0.67$ within that resource but collapse on the newly extracted chemistry ($ρ= 0.12$), while models trained on the new compounds transfer back successfully ($ρ= 0.80$). We conclude that the current composition of the published records, and not dataset size, is limiting the construction of more generalizable models, and we outline what would need to change in reporting practices for better data-driven permeability models.