PhAME: Phenotype-Aware Molecular Editing via Latent Diffusion

2026-05-27

AI Summary

This work addresses the challenge in small-molecule drug discovery of simultaneously optimizing target phenotypic outcomes while preserving structural similarity to a seed compound. The authors formulate molecular optimization as an editing task in the latent space of a pretrained graph variational autoencoder (GVAE) and introduce a latent diffusion model conditioned on phenotypic profiles to enable phenotype-guided generation. A novel dual-scale classifier-free guidance mechanism is proposed to independently modulate the strength of phenotypic guidance and the degree of structural similarity to the seed molecule. Evaluated across multiple benchmark tasks, the method achieves state-of-the-art performance, enabling precise co-control over molecular properties and structure while maintaining high chemical validity and novelty.

Abstract

Small-molecule drug discovery requires simultaneous optimization of numerous properties of candidate molecules. These properties can be investigated through the analysis of high-dimensional biological signatures, such as cell morphology and transcriptomic perturbations, which provide a rich perspective on the underlying biological mechanisms. However, existing generative methods, which use those signatures for optimization, fail to meet two key requirements: providing precise guidance toward desired phenotypic signatures while maintaining structural proximity to a known hit. We introduce PhAME (Phenotype-Aware Molecular Editing), a latent diffusion framework that overcomes this challenge by recasting molecular optimization as editing in the latent space of a pretrained graph-based VAE. Our central contribution is a compositional classifier-free guidance scheme with two independent scales, one for the phenotype-conditioning and one for similarity to the seed structure, allowing practitioners to control the tradeoff between these two objectives. Empirical evaluations across diverse benchmarks, including docking score optimization and multimodal phenotypic generation, demonstrate that PhAME achieves state-of-the-art results while maintaining high chemical validity and novelty.

Problem

• molecular optimization
• phenotypic signatures
• structural proximity
• small-molecule drug discovery
• biological mechanisms

Innovation

• latent diffusion
• phenotype-aware editing
• compositional classifier-free guidance
• molecular optimization
• graph-based VAE