The protein folding problem must ultimately be solved on all length scales from the atomic up through a hierarchy of complicated structures. By analyzing the stability of the folding process using physics and mathematics, this paper shows that featur
es without length scales, i.e. topological features, are potentially of central importance. Topology is a natural mathematical tool for the study of shape and we avail ourselves of that tool to examine the relationship between the amino acid sequence and the shapes of protein molecules. We apply what we learn to conjectures about their biological evolution.
Processes that proceed reliably from a variety of initial conditions to a unique final form, regardless of moderately changing conditions, are of obvious importance in biophysics. Protein folding is a case in point. We show that the action principle
can be applied directly to study the stability of biological processes. The action principle in classical physics starts with the first variation of the action and leads immediately to the equations of motion. The second variation of the action leads in a natural way to powerful theorems that provide quantitative treatment of stability and focusing and also explain how some very complex processes can behave as though some seemingly important forces drop out. We first apply these ideas to the non-equilibrium states involved in two-state folding. We treat torsional waves and use the action principle to talk about critical points in the dynamics. For some proteins the theory resembles TST. We reach several quantitative and qualitative conclusions. Besides giving an explanation of why TST often works in folding, we find that the apparent smoothness of the energy funnel is a natural consequence of the putative critical points in the dynamics. These ideas also explain why biological proteins fold to unique states and random polymers do not. The insensitivity to perturbations which follows from the presence of critical points explains how folding to a unique shape occurs in the presence of dilute denaturing agents in spite of the fact that those agents disrupt the folded structure of the native state. This paper contributes to the theoretical armamentarium by directing attention to the logical progression from first physical principles to the stability theorems related to catastrophe theory as applied to folding. This can potentially have the same success in biophysics as it has enjoyed in optics.
We adopt the point of view that analysis of the stability of the protein folding process is central to understanding the underlying physics of folding. Stability of the folding process means that many perturbations do not disrupt the progress from th
e random coil to the native state. In this paper we explore the stability of folding using established methods from physics and mathematics. Our result is a preliminary theory of the physics of folding. We suggest some tests of these ideas using folding simulations. We begin by supposing that folding events are related in some way to mechanical waves on the molecule. We adopt an analytical approach to the physics which was pioneered by M.V. Berry, (in another context), based upon mathematics developed mainly by R. Thom and V.I. Arnold. We find that the stability of the folding process can be understood in terms of structures known as caustics, which occur in many kinds of wave phenomena. The picture that emerges is that natural selection has given us a set of protein molecules which have mechanical waves that propagate according to several mathematically specific restrictions. Successful simulations of folding can be used to test and constrain these wave motions. With some additional assumptions the theory explains or is consistent with a number of experimental facts about folding. We emphasize that this wave-based approach is fundamentally different from energy-based approaches.
The energy for protein folding arises from multiple sources and is not large in total. In spite of the many specific successes of energy landscape and other approaches, there still seems to be some missing guiding factor that explains how energy from
diverse small sources can drive a complex molecule to a unique state. We explore the possibility that the missing factor is in the geometry. A comparison of folding with other physical phenomena, together with analytic modeling of a molecule, led us to analyze the physics of optical caustic formation and of folding behavior side-by-side. The physics of folding and caustics is ostensibly very different but there are several strong parallels. This comparison emphasizes the mathematical similarity and also identifies differences. Since the 1970s, the physics of optical caustics has been developed to a very high degree of mathematical sophistication using catastrophe theory. That kind of quantitative application of catastrophe theory has not previously been applied to folding nor have the points of similarity with optics been identified or exploited. A putative underlying physical link between caustics and folding is a torsion wave of non-constant wave speed, propagating on the dihedral angles and $Psi$ found in an analytical model of the molecule. Regardless of whether we have correctly identified an underlying link, the analogy between caustic formation and folding is strong and the parallels (and differences) in the physics are useful.
A geometric analysis of protein folding, which complements many of the models in the literature, is presented. We examine the process from unfolded strand to the point where the strand becomes self-interacting. A central question is how it is possibl
e that so many initial configurations proceed to fold to a unique final configuration. We put energy and dynamical considerations temporarily aside and focus upon the geometry alone. We parameterize the structure of an idealized protein using the concept of a ribbon from differential geometry. The deformation of the ribbon is described by introducing a generic twisting Ansatz. The folding process in this picture entails a change in shape guided by the local amino acid geometry. The theory is reparamaterization invariant from the start, so the final shape is independent of folding time. We develop differential equations for the changing shape. For some parameter ranges, a sine-Gordon torsion soliton is found. This purely geometric waveform has properties similar to dynamical solitons. Namely: A threshold distortion of the molecule is required to initiate the soliton, after which, small additional distortions do not change the waveform. In this analysis, the soliton twists the molecule until bonds form. The analysis reveals a quantitative relationship between the geometry of the amino acids and the folded form.
In a ring of s-wave superconducting material the magnetic flux is quantized in units of $Phi_0 = frac{h}{2e}$. It is well known from the theory of Josephson junctions that if the ring is interrupted with a piece of d-wave material, then the flux is q
uantized in one-half of those units due to a additional phase shift of $pi$. We reinterpret this phenomenon in terms of geometric phase. We consider an idealized hetero-junction superconductor with pure s-wave and pure d-wave electron pairs. We find, for this idealized configuration, that the phase shift of $pi$ follows from the discontinuity in the geometric phase and is thus a fundamental consequence of quantum mechanics.
