#32 · Tools & infrastructure

Web servers now predict RNA structure, protein binding, and DNA variant effects

The dominant thread is the packaging of structural and sequence-context features into freely accessible web servers for predicting molecular function. ShapeRNA integrates RNA secondary structure prediction, SHAPE/DMS-guided modeling, and ensemble inference with structure-aware regulatory analysis. RNAanalyzer3 provides RNA motif discovery and motif-in-context analysis, linking hits to Rfam and miRBase with batch processing. SPSignal combines sequence motifs with solvent accessibility, disorder, and structural context to predict NLS and NES signals. xBind uses ESM-2 embeddings and symmetry-aware graph neural networks to predict protein-protein and protein-DNA binding sites. FABIAN-variant 2026 addresses how DNA variants alter transcription factor binding. The set is heterogeneous; only the two RNA tools share an explicit RNA structural framework. What appears new is the integration of experimental probing data and language-model embeddings into servers, though the items do not report benchmark comparisons.