Structure Alignment-driven Cross-Graph Modeling for Functional RNA Design

Abstract RNAs are critical for biological processes, with their biological functions closely tied to their three-dimensional structures. RNA inverse folding, the design of RNA sequences that fold into target 3D structures, is a complex challenge due to the dynamic and unstable nature of RNA structures. Motivated by evolutionary conservation concepts from structure prediction, we present a novel RNA design approach AlignIF, which leverages multiple structure alignment (MStA) with cross-graph modeling to capture evolutionarily conserved structural patterns at the structural level, facilitating RNA sequence design. AlignIF outperforms existing state-of-the-art methods by a large margin in key metrics, including sequence recovery, perplexity, and foldability. Notably, it enables the design of entire RNA families rather than being restricted to recapitulating native sequences. Furthermore, AlignIF effectively generates functional RNA fluorescent aptamers and self-cleaving ribozymes, as experimentally validated by their respective fluorescent signals or cleavage activity. Importantly, two of the ten designed aptamers show enhanced fluorescence compared to the wild-type aptamer due to the increased fluorophore binding capacity, and other two exhibit improved binding affinity toward the target molecule, highlighting the potential of AlignIF for engineering novel functional RNAs.