A Multi-Level Contrastive Training Protocol for Hierarchical Generative Models
3D pride without 2D prejudice: Bias-controlled multi-level generative models for structure-based ligand design
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.The framework leverages the large data resources available for 2d generative modelling with datasets of ligand-protein complexes.Hierarchical generative models that are topologically unbiased, explainable and customizable are produced by deconvolving the generativeposterior into chemical, topological and structural context factors.We show how, by deconvolving the generativeposterior 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.
