Accelerated Simulations\nReveal Physicochemical Factors\nGoverning Stability and Composition of RNA Clusters

Under certain conditions,\nRNA repeat sequences phase\nseparate,\nyielding protein-free biomolecular condensates. Importantly, RNA repeat\nsequences have also been implicated in neurological disorders, such\nas Huntington’s disease. Thus, mapping repeat sequences to\ntheir phase behavior, functions, and dysfunctions is an active area\nof research. However, despite several advances, it remains challenging\nto characterize the RNA phase behavior at a submolecular resolution.\nHere, we have implemented a residue-resolution coarse-grained model\nin LAMMPSthat incorporates both the RNA sequence and structureto\nstudy the clustering propensities of protein-free RNA systems. Importantly,\nwe achieve a multifold speedup in the simulation time compared to\nprevious work. Leveraging this efficiency, we study the clustering\npropensity of all 20 nonredundant trinucleotide repeat sequences.\nOur results align with findings from experiments, emphasizing that\ncanonical base-pairing and G–U wobble pairs play dominant roles\nin regulating cluster formation of RNA repeat sequences. Strikingly,\nwe find strong entropic contributions to the stability and composition\nof RNA clusters, which is demonstrated for single-component RNA systems\nas well as binary mixtures of trinucleotide repeats. Additionally,\nwe investigate the clustering behaviors of trinucleotide (odd) repeats\nand their quadranucleotide (even) counterparts. We observe that odd\nrepeats exhibit stronger clustering tendencies, attributed to the\npresence of consecutive base pairs in their sequences that are disrupted\nin even repeat sequences. Altogether, our work extends the set of\ncomputational tools for probing RNA cluster formation at submolecular\nresolution and uncovers physicochemical principles that govern the\nstability and composition of the resulting clusters.