COCONUT: An analysis of coiled-coil regions in proteins
Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Motivation
The molecular rules determine the strength and orientation (parallel or antiparallel) of interacting coiled-coil helices in protein-protein interactions. Interpreting these rules is crucial for identifying novel protein-protein interactions, designing competitive binders, and constructing large assemblies containing coiled-coil domains. This study establishes the molecular principles that dictate the strength and orientation of coiled-coil interactions, providing insights relevant to these applications.
Results
We examined how hydrophobic contacts determine structural specificity within coiled-coil dimers. Our analysis revealed that the hydrophobic core densities differ between parallel and antiparallel dimer confirmations, highlighting their importance in stabilizing different structural arrangements. We developed CO iled- CO il a N alysis UT ility (COCONUT), a computational platform with machine learning models, validated for predictive capabilities in various applications. Using COCONUT’s pipeline for coiled-coil analysis and modeling, we predicted the orientation of substitution-sensitive coiled-coil dimer, identified residue pairings in non-canonical coiled-coil heterodimer, and constructed n-stranded coiled-coil model. These results demonstrate COCONUT’s utility as a computational framework for interpreting and modeling coiled-coil structures.
Availability and implementation
COCONUT is an open-source and free Python package available here https://github.com/neeleshsoni21/COCONUT . The documentation is available in the source code and here: https://neeleshsoni21.github.io/COCONUT/
