Search papers, labs, and topics across Lattice.
The paper introduces REUSE, a novel method for generating dual-target molecules using a pre-trained single-target diffusion model without retraining or modifying the diffusion process. REUSE frames dual-target molecule generation as a constrained multi-objective optimization problem within the input space of the frozen diffusion model. Results demonstrate that REUSE significantly improves dual-target affinity and balance compared to methods that modify the diffusion process, while maintaining molecular quality.
You can generate high-quality molecules that hit two targets at once, without the cost of retraining your generative model.
Designing a single molecule that modulates two targets is a promising strategy for polypharmacology, but it remains substantially harder than standard single-target generation because one candidate must satisfy two binding requirements while preserving drug-likeness and synthesizability. Existing dual-target generative methods typically introduce dual-target capability by either retraining the generator or intervening in the diffusion process during sampling. The former can be costly and difficult to stabilize when dual-target supervision is sparse, while the latter may be sensitive to denoising-time target balancing and competing update directions. These limitations motivate a generator-preserving alternative that keeps the pretrained prior intact: can dual-target candidates instead be recovered from the input space of a frozen single-target diffusion model, without modifying its parameters or denoising dynamics? We formulate this task as a constrained multi-objective optimization problem and propose REUSE, a hierarchical evolutionary input-space search framework that combines pair-conditioned exploration with structured multi-stage selection to enforce dual-target affinity, chemical quality, and diversity. Experiments show that, compared with methods that modify the diffusion process, REUSE consistently improves dual-target affinity and balance, achieving a 20.9-percentage-point gain in Dual High Affinity over the strongest prior baseline while maintaining competitive molecular quality.
