Genotype-Conditioned Molecular Generation via Evidence-Grounded Multi-Objective Latent Perturbation in Diffusion Models

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Nogueira, Brenda, Gonzalez-Montiel, Gisela A., Chawla, Nitesh V., Moniz, Nuno
Formato: Preprint
Publicado: 2026
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866918534434521088
author Nogueira, Brenda
Gonzalez-Montiel, Gisela A.
Chawla, Nitesh V.
Moniz, Nuno
author_facet Nogueira, Brenda
Gonzalez-Montiel, Gisela A.
Chawla, Nitesh V.
Moniz, Nuno
contents Developing effective anticancer therapeutics remains challenging due to tumor heterogeneity and the absence of well-defined molecular targets across cancer subtypes. Generative models conditioned on cancer genotypes offer a promising avenue for personalized drug discovery, yet existing approaches lack explicit optimization for simultaneous sensitivity, synthesizability, and mechanistic binding plausibility. We present a latent-space optimization approach for a pretrained genotype-to-drug diffusion model, introducing a learnable perturbation over the molecular latent space optimized via gradient ascent to maximize a composite reward combining predicted drug sensitivity (AUC), drug-likeness (QED), and synthetic accessibility (SAS). Critically, biological realism is enforced by grounding both reward design and evaluation in experimentally-derived cancer cell line data and validated pharmacologic signals, anchoring candidate generation in real-world clinical evidence. Mechanistic consistency plausibility is further assessed by a multi-agent LLM pipeline grounded in the diffusion model's attention mechanism. Experiments across 15 cancer cell lines from three held-out evaluation sets demonstrate consistent and noticeable improvements over competing baselines in sensitivity, drug-likeness, synthesizability, and chemical validity.
format Preprint
id arxiv_https___arxiv_org_abs_2606_01461
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Genotype-Conditioned Molecular Generation via Evidence-Grounded Multi-Objective Latent Perturbation in Diffusion Models
Nogueira, Brenda
Gonzalez-Montiel, Gisela A.
Chawla, Nitesh V.
Moniz, Nuno
Machine Learning
Multiagent Systems
Developing effective anticancer therapeutics remains challenging due to tumor heterogeneity and the absence of well-defined molecular targets across cancer subtypes. Generative models conditioned on cancer genotypes offer a promising avenue for personalized drug discovery, yet existing approaches lack explicit optimization for simultaneous sensitivity, synthesizability, and mechanistic binding plausibility. We present a latent-space optimization approach for a pretrained genotype-to-drug diffusion model, introducing a learnable perturbation over the molecular latent space optimized via gradient ascent to maximize a composite reward combining predicted drug sensitivity (AUC), drug-likeness (QED), and synthetic accessibility (SAS). Critically, biological realism is enforced by grounding both reward design and evaluation in experimentally-derived cancer cell line data and validated pharmacologic signals, anchoring candidate generation in real-world clinical evidence. Mechanistic consistency plausibility is further assessed by a multi-agent LLM pipeline grounded in the diffusion model's attention mechanism. Experiments across 15 cancer cell lines from three held-out evaluation sets demonstrate consistent and noticeable improvements over competing baselines in sensitivity, drug-likeness, synthesizability, and chemical validity.
title Genotype-Conditioned Molecular Generation via Evidence-Grounded Multi-Objective Latent Perturbation in Diffusion Models
topic Machine Learning
Multiagent Systems
url https://arxiv.org/abs/2606.01461