Morpheus-3D: Structural Diversity-Guided Detection and Localization of Protein Fold Switching
Kuniyil, S.; Subramanian, V.; Arun, A.; Lakshmanan, A.; Sekhar, A.; Srivastava, A.
Show abstract
Proteins that reversibly adopt multiple stable folds challenge the classical sequence-structure paradigm, yet their discovery remains limited because fold switching is difficult to detect experimentally and current computational methods fail to resolve the underlying conformationally plastic regions. Here we present Morpheus-3D, a sequence-based framework that quantifies residue-level tertiary structural diversity using entropy profiles derived from the Foldseek 3Di structural alphabet. By capturing variation in tertiary interaction environments rather than secondary structure alone, Morpheus-3D identifies fold-switching proteins while simultaneously localizing the sequence regions responsible for structural transitions. The framework outperforms existing predictors, accurately recovers experimentally characterized switching regions, generalizes to recently discovered natural and engineered fold-switching proteins absent from training, and detects conformational plasticity inaccessible to secondary-structure-based approaches. Application to 57 representative proteomes reveals that fold-switching potential is widespread but enriched in regulatory, pathogenic, and environmentally adaptive lineages. Integration with ancestral sequence reconstruction further un-covers evolutionary trajectories through which conformational plasticity emerges. To make these predictions directly accessible, we implemented Morpheus-3D as an interactive web platform (https://morpheus.slicearrow.com/), in which per-residue entropy profiles, sequence and three-dimensional structure are displayed together and respond as one, allowing predicted fold-switching regions to be mapped onto the structure and exported for downstream analysis. Morpheus-3D provides a scalable framework for discovering metamorphic proteins and investigating the origins, mechanisms, and evolution of structural plasticity directly from sequence.
Matching journals
The top 4 journals account for 50% of the predicted probability mass.
Similar papers in this journal
Similar papers in this journal
- ConforFold Recovers Alternative Protein Conformations Beyond MSA Subsampling 96%
- COLLAPSE: A representation learning framework for identification and characterization of protein structural sites 95%
- The amino acid sequence determines protein abundance through its conformational stability and reduced synthesis cost. 94%
Similar papers in this journal
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.