arXiv:2605.28226v1 Announce Type: new 
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.

تم تقديم إطار جديد يسمى PhAME (تحرير الجزيئات الواعي بالظاهرة) يستخدم الانتشار الكامن لتحسين اكتشاف الأدوية الجزيئية الصغيرة من خلال تمكين التوجيه الدقيق نحو التوقيعات الظاهرة المرغوبة مع الحفاظ على التشابه الهيكلي مع المركبات المعروفة.

Se ha introducido un nuevo marco llamado PhAME (Edición Molecular Consciente del Fenotipo), que utiliza la difusión latente para optimizar el descubrimiento de fármacos de pequeñas moléculas, permitiendo una orientación precisa hacia firmas fenotípicas deseadas mientras se mantiene la similitud estructural con compuestos conocidos.

Un nouveau cadre nommé PhAME (Édition Moléculaire Sensible au Phénotype) a été introduit, utilisant la diffusion latente pour optimiser la découverte de médicaments à petites molécules en permettant une orientation précise vers des signatures phénotypiques souhaitées tout en maintenant une similarité structurelle avec des composés connus.

A new framework named PhAME (Phenotype-Aware Molecular Editing) has been introduced, utilizing latent diffusion to optimize small-molecule drug discovery by enabling precise guidance towards desired phenotypic signatures while maintaining structural similarity to known compounds.