Memory in models of molecular assembly
File(s)
Author(s)
Whitby, Samuel Anthony Isaac
Type
Thesis or dissertation
Abstract
We explore how the existence of frustrated kinetics in molecular interactions influences the emergent properties of the structures that they assemble into. These structures range in scale from folded shapes made from single proteins, through intricate, structurally defined complexes that contain hundreds of different components, to entire phase-separated droplets that feature a vast number of weakly interacting entities within a freely-flowing fluid.
A lattice gas model for biomolecular condensates is proposed, with coupling between sites allowed to vary through time to account for diffusion over a rugged energy landscape of inter- acting states. The model demonstrates gradual ageing within entire droplets and clarifies the necessary conditions for heterogeneous structures to form. By describing greater frustration for interactions between particles, a system with glassy kinetics is paradoxically shown to achieve faster rates of equilibration.
A lattice model of stoichiometrically defined biomolecular complexes is introduced that has a Go ̄-type energy with long-range repulsion and evolves according to dynamics based on the Virtual Move Monte Carlo method. Systems with temporal schedules and spatial gradients of chemical parameters are considered with this model, both showing enhanced rates of complex assembly that resembles the effect of annealing schedules. The model provides insight into many laboratory reconstitution protocols and suggests a role for chemical gradients within condensates such as the nucleolus that is involved in ribogenesis.
The inverse problem of how a biomolecule can be engineered to allow its assembly status to reflect its history is then considered. A model is introduced for protein folding that traverses hierarchical energy landscape that require a defined temporal sequence of binding strengths, which allows it to resemble a finite state machine. By considering the pH dependence of amino acid residues, proteins are conceptualised as hidden-layer perceptrons capable of computational tasks that involve memory storage and signal classification.
A lattice gas model for biomolecular condensates is proposed, with coupling between sites allowed to vary through time to account for diffusion over a rugged energy landscape of inter- acting states. The model demonstrates gradual ageing within entire droplets and clarifies the necessary conditions for heterogeneous structures to form. By describing greater frustration for interactions between particles, a system with glassy kinetics is paradoxically shown to achieve faster rates of equilibration.
A lattice model of stoichiometrically defined biomolecular complexes is introduced that has a Go ̄-type energy with long-range repulsion and evolves according to dynamics based on the Virtual Move Monte Carlo method. Systems with temporal schedules and spatial gradients of chemical parameters are considered with this model, both showing enhanced rates of complex assembly that resembles the effect of annealing schedules. The model provides insight into many laboratory reconstitution protocols and suggests a role for chemical gradients within condensates such as the nucleolus that is involved in ribogenesis.
The inverse problem of how a biomolecule can be engineered to allow its assembly status to reflect its history is then considered. A model is introduced for protein folding that traverses hierarchical energy landscape that require a defined temporal sequence of binding strengths, which allows it to resemble a finite state machine. By considering the pH dependence of amino acid residues, proteins are conceptualised as hidden-layer perceptrons capable of computational tasks that involve memory storage and signal classification.
Version
Open Access
Date Issued
2025-12-04
Date Awarded
2026-07-01
Copyright Statement
Attribution-NonCommercial-ShareAlike 4.0 International Licence (CC BY NC-SA)
Advisor
Lee, Chiu Fan
Sponsor
Imperial College London
Publisher Department
Department of Bioengineering
Publisher Institution
Imperial College London
Qualification Level
Doctoral
Qualification Name
Doctor of Philosophy (PhD)