Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
MORAD is an online reinforcement learning framework for RNA inverse design that trains a single backbone-conditioned policy to optimize multiple, heterogeneous objectives. By jointly learning tertiary geometry, base pairing, ensemble behavior, and sequence composition via a weighted geometric mean of bounded desirability scores, MORAD significantly outperforms existing methods across 10 of 11 evaluation metrics. This approach establishes a new state of the art in structure-conditioned RNA inverse design, improving key metrics like GDT-TS by 16.0% and pairing MCC by 19.4%.
A single RL policy can now design RNA sequences for diverse backbones, achieving state-of-the-art performance across structural accuracy, pairing, and thermodynamic quality by optimizing six distinct objectives.
We introduce MORAD, an online reinforcement learning framework for backbone-conditioned RNA diffusion models that trains a single shared policy across diverse target backbones. Structure-conditioned RNA inverse design seeks nucleotide sequences compatible with a target three-dimensional backbone, yet existing reinforcement learning approaches either optimize a separate policy for each target backbone or rely on offline preference optimization, leaving online post-training of a reusable backbone-conditioned model largely unexplored. MORAD combines backbone-conditioned group sampling, group-relative reward normalization, and a forward-process diffusion objective. Beyond existing RL rewards that primarily target tertiary structural quality, MORAD jointly optimizes six measurements spanning tertiary geometry, base pairing, ensemble behavior, and sequence composition. To aggregate these heterogeneous objectives despite differences in scale and optimization direction, MORAD maps each measurement to a bounded desirability score and combines them with a weighted geometric mean. Compared with a tertiary-structure-only reward, this multi-objective formulation improves 10 of the 11 evaluation metrics that have a preferred direction. Applied to pretrained RIDE, MORAD improves GDT-TS by 16.0% (from 0.3300 to 0.3827) and pairing MCC by 19.4% (from 0.6110 to 0.7295), while reducing normalized ensemble defect by 27.5% (from 0.3906 to 0.2830). On the test set, MORAD outperforms RiboDiffusion, gRNAde, and RDesign across tertiary structural accuracy, secondary-structure pairing, and thermodynamic quality, establishing a new state of the art in structure-conditioned RNA inverse design.