# De novo PROTAC design using graph-based deep generative models

arXiv: https://arxiv.org/abs/2211.02660 · HTML: https://mlsanity.com/paper/2211.02660/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 2 of its 35 references that are in the map.
- First posted on arXiv 4 Nov 2022.

## Builds on (most cited first, 2 of 2)

- [Sample Efficiency Matters: A Benchmark for Practical Molecular Optimization](https://mlsanity.com/paper/2206.12411/) (2022)
- [Deep learning for molecular design - a review of the state of the art](https://mlsanity.com/paper/1903.04388/) (2019)

## Built on by (most cited first, 3 of 3)

- [Reinforcement Learning-Driven Linker Design via Fast Attention-based Point Cloud Alignment](https://mlsanity.com/paper/2306.08166/) (2023)
- [MoleCLUEs: Molecular Conformers Maximally In-Distribution for Predictive Models](https://mlsanity.com/paper/2306.11681/) (2023)
- [CONFIDE: Hallucination Assessment for Reliable Biomolecular Structure Prediction and Design](https://mlsanity.com/paper/2512.02033/) (2025)

## Abstract (arXiv)

PROteolysis TArgeting Chimeras (PROTACs) are an emerging therapeutic modality for degrading a protein of interest (POI) by marking it for degradation by the proteasome. Recent developments in artificial intelligence (AI) suggest that deep generative models can assist with the de novo design of molecules with desired properties, and their application to PROTAC design remains largely unexplored. We show that a graph-based generative model can be used to propose novel PROTAC-like structures from empty graphs. Our model can be guided towards the generation of large molecules (30--140 heavy atoms) predicted to degrade a POI through policy-gradient reinforcement learning (RL). Rewards during RL are applied using a boosted tree surrogate model that predicts a molecule's degradation potential for each POI. Using this approach, we steer the generative model towards compounds with higher likelihoods of predicted degradation activity. Despite being trained on sparse public data, the generative model proposes molecules with substructures found in known degraders. After fine-tuning, predicted activity against a challenging POI increases from 50% to >80% with near-perfect chemical validity for sampled compounds, suggesting this is a promising approach for the optimization of large, PROTAC-like molecules for targeted protein degradation.