Excited to announce a new collaborative preprint about a structural concept 'local stability compensation'. This states structurally important motifs must be flanked by more stable helices. We observe this effect natural occurring RNAs and experimentally evaluate it. www.biorxiv.org/content/10.1...

Most significantly, we discover that DMS reactivity correlates strongly with atomic distances in non-canonical base pairs. These quantitative relationships demonstrate that DMS chemical mapping data encodes detailed information about RNA 3D structure.

Most significantly, we discover that DMS reactivity correlates strongly with atomic distances in non-canonical base pairs. These quantitative relationships demonstrate that DMS chemical mapping data encodes detailed information about RNA 3D structure.

We find that 11% of non-Watson-Crick nucleotides show protection from DMS similar to Watson-Crick pairs. This protection stems from hydrogen bonding and reduced solvent accessibility. Sequence context can alter reactivity up to 100-fold in specific non-canonical pairs.

We analyzed flanking WC pairs and found structural features that determine their DMS reactivity. A-U pairs are 19-fold more reactive than G-C pairs, purine neighbors increase reactivity, and junction asymmetry correlates with higher reactivity.

Analysis of our comprehensive dataset reveals DMS reactivity exists on a continuous spectrum rather than discrete states. We observe ~10% overlap between Watson-Crick and non-Watson-Crick nucleotides, demonstrating that simple reactivity thresholds cannot reliably determine base-pairing status.

To correlate DMS reactivity with RNA structure, we built a massive library of 7,500 RNA constructs containing multiple junctions with known 3D structures. Our measurements are highly reproducible (R²=0.99), span four orders of magnitude, and reveal that RNA motifs have unique DMS profiles.