# 3D pride without 2D prejudice: Bias-controlled multi-level generative models for structure-based ligand design

arXiv: https://arxiv.org/abs/2204.10663 · HTML: https://mlsanity.com/paper/2204.10663/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 13 of its 52 references that are in the map.
- First posted on arXiv 22 Apr 2022.

## Builds on (most cited first, 13 of 13)

- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [Fast Graph Representation Learning with PyTorch Geometric](https://mlsanity.com/paper/1903.02428/) (2019)
- [Graph Contrastive Learning with Augmentations](https://mlsanity.com/paper/2010.13902/) (2020)
- [Automatic chemical design using a data-driven continuous representation of molecules](https://mlsanity.com/paper/1610.02415/) (2016)
- [Keystroke Dynamics as Part of Lifelogging](https://mlsanity.com/paper/2010.10144/) (2020)
- [GuacaMol: Benchmarking Models for De Novo Molecular Design](https://mlsanity.com/paper/1811.09621/) (2018)
- [Molecular De Novo Design through Deep Reinforcement Learning](https://mlsanity.com/paper/1704.07555/) (2017)
- [Recommendations as Treatments: Debiasing Learning and Evaluation](https://mlsanity.com/paper/1602.05352/) (2016)
- [Deep Reinforcement Learning for De-Novo Drug Design](https://mlsanity.com/paper/1711.10907/) (2017)
- [Inverse design of 3d molecular structures with conditional generative neural networks](https://mlsanity.com/paper/2109.04824/) (2021)
- [Symmetry-Aware Actor-Critic for 3D Molecular Design](https://mlsanity.com/paper/2011.12747/) (2020)
- [Contrastive Learning for Debiased Candidate Generation in Large-Scale Recommender Systems](https://mlsanity.com/paper/2005.12964/) (2020)
- [A Deep Generative Model for Molecule Optimization via One Fragment Modification](https://mlsanity.com/paper/2012.04231/) (2020)

## Built on by (most cited first, 1 of 1)

- [Structure-based drug discovery with deep learning](https://mlsanity.com/paper/2212.13295/) (2022)

## Abstract (arXiv)

Generative models for structure-based molecular design hold significant promise for drug discovery, with the potential to speed up the hit-to-lead development cycle, while improving the quality of drug candidates and reducing costs. Data sparsity and bias are, however, two main roadblocks to the development of 3D-aware models. Here we propose a first-in-kind training protocol based on multi-level contrastive learning for improved bias control and data efficiency. The framework leverages the large data resources available for 2D generative modelling with datasets of ligand-protein complexes. The result are hierarchical generative models that are topologically unbiased, explainable and customizable. We show how, by deconvolving the generative posterior into chemical, topological and structural context factors, we not only avoid common pitfalls in the design and evaluation of generative models, but furthermore gain detailed insight into the generative process itself. This improved transparency significantly aids method development, besides allowing fine-grained control over novelty vs familiarity.