DOE OSTI · 3451234
Deep learning–driven rare-event sampling reveals an Ala7/d-Ala8 conformational hinge governing amphiphilic switching in cyclosporine A
Abstract
Passive membrane permeability of N-methylated macrocyclic peptides remains poorly understood because high-energy conformational intermediates are rarely sampled by conventional methods. Using the deep learning–driven rare-event sampling framework ES-CRUMPLE, we show that cyclosporine A accesses transient amphiphilic states via a localized conformational hinge at Ala7/d-Ala8, revealing a kinetic design handle that is invisible to population-based analyses.
Keep this discovery
Explore connections, maps & timelines
Lindberg, Edward T [University of Tennessee, Knoxville (UTK)], Hsu, Darren [NVIDIA, Santa Clara, CA], Ngo, Van [ORNL] (ORCID:0000000183501203), Do, Thanh D [University of Tennessee, Knoxville (UTK)]. 2026-08-01. Deep learning–driven rare-event sampling reveals an Ala7/d-Ala8 conformational hinge governing amphiphilic switching in cyclosporine A. https://doi.org/10.1039/d6cc04209k
Cite the original work for its findings. Save a collection to share your selection of sources.