Learning interpretable kinetic models for biomolecular interaction networks
Many cellular machines operate through weak, transient, and multivalent interactions whose functional states are governed by recurring interaction patterns rather than persistent molecular geometries. Here, we introduce interaction-based Markov state models (iMSMs), which construct interpretable kinetic models using unsupervised clustering of time-averaged, identity-resolved interaction…
Cellular machines function using weak, temporary, and multivalent interactions, with functional states determined by recurring interaction patterns instead of stable molecular structures. Researchers have developed interaction-based Markov state models (iMSMs), which create interpretable kinetic models through unsupervised clustering of time-averaged, identity-resolved interaction distributions surrounding a specific molecule.
These iMSMs were tested at two molecular levels of nucleocytoplasmic transport and were able to identify graded interaction states ranging from strong, partial, to weak engagement.
During pore transport, partially engaged states facilitated swifter transitions to disengagement compared to tightly bound states. The networks uncovered transport pathways, interaction hubs, obstacles, and kinetic commitment. At a more detailed resolution, FG motifs exchanged contacts while remaining affiliated, reproducing the well-known slide-and-exchange dynamics.
When compared to direct transport-event counting, iMSMs were able to accurately reproduce free energies, permeabilities, and multi-step kinetics, and achieved nearly four times faster permeability convergence from truncated trajectories. Overall, iMSMs succeeded in connecting swiftly exchanging contacts to graded interaction states and their kinetics across multiple molecular scales.
Written by urgent.news from bioRxiv's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
