We are inclined to separate the brain and the muscle — the brain does the considering; the muscle does the doing. The brain takes in complex information concerning the world and makes decisions, and the muscle merely executes. This has also shaped how we take into consideration a single cell; some molecules inside cells are seen as ‘thinkers’ that soak up information concerning the chemical environment and judge what the cell must do for survival; individually, other molecules are seen because the ‘muscle,’ constructing structures needed for survival.
But a brand new study shows how the molecules that construct structures, i.e, the muscle, can themselves do each the considering and the doing. The study, by scientists with the University of Chicago, California Institute of Technology, and Maynooth University, was published Jan. 17 in Nature, and should suggest avenues for brand spanking new ways to take into consideration computation using the principles of physics.
“We show that a natural molecular process — nucleation — that has been studied as a ‘muscle’ for a very long time can do complex calculations that rival an easy neural network,” said UChicago Assoc. Prof. Arvind Murugan, one in every of the 2 senior co-authors on the paper. “It’s a capability hidden in plain sight — the ‘doing’ molecules may also do the ‘considering.’ Evolution can exploit this fact in cells to get more done with fewer parts, with less energy and greater robustness.”
Considering using physics
To survive, cells need to acknowledge the environment they’re in and respond accordingly. For instance, some combos of molecules might indicate a time of stress that requires hunkering down, while other combos of molecules might indicate a time of plenty. However the difference between these molecular signals might be subtle — different environments might involve the identical molecules but in several proportions.
Constantine Evans, the lead writer of the study, explained that it’s a bit like walking right into a house and smelling freshly baked cookies, versus smelling burning rubber. “Your brain would alter your behavior depending on you sensing different combos of odorful chemicals,” he said. “We got down to ask if just the physics of a molecular system can do the identical, despite not having a brain of any kind.”
The standard view has been that cells might have the option to sense and respond in this fashion using molecular circuits that conceptually resemble the electronic circuits in your laptop; some molecules sense the quantity of salt and acid within the environment, other molecules make a choice on what to do, and at last ‘muscle’ molecules might perform an motion in response, like constructing an internal protective structure or a pump to remove unwanted molecules.
Murugan and his colleagues desired to explore an alternate idea: that every one of those tasks — sensing, decision making, response — might be completed in a single step by the physics inherent to the ‘muscle’ molecules that construct a structure.
They did so by working with the principle of “phase transitions.” Consider a glass of water freezing when it hits 32F; first, a bit fragment of ice ‘nucleates,’ after which grows out until the entire glass of water is frozen.
On the face of it, these initial steps within the act of freezing — called ‘nucleation’ in physics — doesn’t resemble ‘considering’. But the brand new study shows that the act of freezing can “recognize” subtly different chemical combos — e.g., the smell of oatmeal raisin cookies versus chocolate chip — and construct different molecular structures in response.
Robustness in experiments
The scientists tested the robustness of ‘phase transitions’-based decision-making using DNA nanotechnology, a field that Erik Winfree (BS’91) helped pioneer. They showed that a mix of molecules would form one in every of three structures depending on what concentrations of molecules were present within the beaker.
“In each case, the molecules got here together to construct different nanometer-scale structures in response to different chemical patterns — except the act of constructing the structure in itself made the choice on what to construct,” Winfree said.
The experiment revealed that this ‘muscle’-based decision making was surprisingly robust and scalable. With relatively easy experiments, the researchers could solve pattern recognition problems involving a few thousand sorts of molecules — nearly a 10-fold larger problem than had been done previously using other approaches that separated ‘brain’ and ‘muscle’ components.
The work points at a brand new view of computation that doesn’t involve designing circuits, but fairly designing what physicists call a phase diagram. For instance, for water, a phase diagram might describe the temperature and pressure conditions through which liquid water will freeze or boil, that are ‘muscle’-like material properties. But this work shows that the phase diagram may also encode ‘considering’ along with ‘doing,’ when scaled as much as complex systems with many alternative sorts of components.
“Physicists have traditionally studied things like a glass of water, which has many molecules, but all of them are similar. But a living cell is stuffed with many alternative sorts of molecules that interact with one another in complex ways,” said coauthor Jackson O’Brien (PhD’21), who was involved within the study as a UChicago graduate student in physics. “This ends in distinct emergent capabilities of multi-component systems.”
The idea on this work drew mathematical analogies between such multi-component systems and the idea of neural networks; the experiments pointed to how these multi-component systems can learn the appropriate computational properties through a physical process, very like the brain learns to associate different smells with different actions.
While the experiments here involved DNA molecules in a test tube, the underlying concepts — nucleation in systems with many sorts of components — applies broadly to many other molecular and physical systems, the authors said.
“DNA lets us experimentally study complex mixtures of 1000’s of sorts of molecules, and systematically understand the impact of what number of sorts of molecules there are and the sorts of interactions they’ve, but the idea is general and will apply to any form of molecule,” explained Winfree.
“We hope this work will spur work to uncover hidden ‘considering’ abilities in other multi-component systems that currently appear to merely be ‘muscles,'” said Murugan.
