A Note on Perspective: I am a patient with Parkinson’s disease who has spent a decade at the intersection of clinical research and biotechnology. I’ve watched billions of dollars and millions of hours vanish into models that continue to fail. This piece isn’t a technical proof; it’s an observation. If biology is the architecture of life, physics are the laws that govern how life plays. We have spent decades trying to understand the scaffolding while ignoring the laws that keep everything upright. It’s time for that to change.
Two weeks ago, I helped arrange for Prof. Alberto Espay to give a talk to an amazing patient group in the San Francisco area called PD-Connect. I assumed he would simply give an update on the synuclein idea that first percolated in the back pages of our book, Brain Fables. While I was delighted to see all the progress he had made since we published it 6 years ago, little did I know that the discussion would turn into a full-fledged scientific debate between Alberto and Profs Malu Tansey and Matt Farrer.
I want to start by commending Malu and Matt for challenging Alberto so openly. It is very rare for patients to get the opportunity to listen to these kinds of conversations but it is important that we are made aware that science is not a straightforward process of discovery. There are bumps on the road that make developing therapies for people in need incredibly difficult.
Here is the talk in full, jump to the 48 minute mark to hear the debate…
There was one thing that Prof. Matt Farrer said around the 53:40 mark that struck me: “You do not need to invoke any of those things (those being the biophysical phase transitions that Alberto was speaking about), it is genetic. There is a direct relationship with gene expression and protein expression in the brains of post-mortem individuals with synuclein triplications. I published those results in 2004. You do not need to invoke phase transitions or any of this biophysical stuff….we do not know how that works in a cell, in a lipid enriched environment and with detergents like alpha-synuclein which is also a chaperone…it’s a very fuzzy area.”
First I want to say that I have a lot of respect for Prof. Matt Farrer. I remember being in his lab at the University of British Columbia in 2017 and watching his eyes light up when I told him I was willing to donate some tissue to help him better understand this disease. He has a palpable and genuine excitement for biology and a real need to help individuals afflicted by the diseases he studies. Also I want to note that his argument applies solely to those with Parkinson’s disease that carry a duplication or triplication of the SNCA gene that encodes for the protein alpha-synuclein. I agree with him that if we can lower the genetic over expression levels at an early enough stage in those individuals, it should delay, perhaps even permanently, their disease. It is a trial that urgently needs to happen.
While mendelian genetics may provide the ‘Monster Truck’ that accelerates Parkinson’s, physics provides the environment in which that truck must function. The distinction between a massive genetic driver and a subtle risk factor becomes secondary when you realize both are bound by the same physical constants. Gravity doesn’t negotiate with a mutation. If we only study the ‘truck’—the biology—we are effectively analyzing engine parts to explain a pile-up caused by an icy road. To truly understand why the system fails, we have to look beyond the machinery and consult the ‘Manual’: the biophysical laws of entropy and energy dissipation that dictate whether a cell stays in motion or grinds to a halt. Physics allows us to predict when the road will ice over, whereas biology only tells us the truck has a heater. *
Measuring post-mortem protein levels is akin to looking at a computer that is turned off and saying, “I see the silicon, therefore I understand the software.” Gene expression is not static, it responds to its environment, immune challenges and age, but it is still a form of cell-type specific hardware. However, disease and thought are dynamic processes. To understand how a brain fails, you must understand the flow of energy and information. You can’t measure a “rhythm” or a “signal” by looking at a dead protein; you need the physics of oscillatory dynamics.
Matt went on to say in the video above: “I agree that there’s two things that are extreme. One is the families (with duplication and triplications of the SNCA gene), they clearly have increased genetic dosage, that (correlates) with age of onset and dose…You don’t need any biophysics in the middle. (The data is) unequivocal. Second, in idiopathic Parkinson’s disease-susceptibility alleles may be modestly correlated with lower levels of gene expression. So you’re right in that sense. And this is the paradox between the two. It is very difficult, and I’ve not been able to get my head around it for 25 plus years.”
I reached out to Dr. Kariem Ezzat, a research scientist at the University of Cincinnati and Stockholm University studying neurodegenerative disease etiology, pathogenesis, and therapeutics to see if he could elaborate: “Matt acknowledges a paradox that he spent years not being able to understand that risk alleles for idiopathic PD appear to marginally lower synuclein expression while gene duplications also lead to PD.”
The paradox mentioned by Matt and Kariem highlights a contradiction that while gene triplication leads to Parkinson’s by increasing protein production, certain risk alleles for the same disease actually appear to lower their production. This might be resolved through biophysics, which explains that “low” measured levels can occur because high protein concentrations facilitate rapid aggregation, effectively locking the proteins away in solid clumps and removing them from the soluble pool being measured. Without the biophysical understanding of “soluble pools” vs. “aggregates,” the genetic data is actually misleading.
Matt, Kariem and I got together on a subsequent call and some emails over the last two weeks to look closely at the data. While Matt originally observed a trend in his lab’s data, when combined with synuclein gene expression from two other public sources the combined results show SNCA disease-susceptibility alleles are NOT correlated with lower levels of gene expression after all.**
The debate between Alberto, Kariem, Malu and Matt illustrates how scientific explanations are not fixed, but may change, as more data accrues, and as scientists come together and start pooling analysis. I believe the exchange serves as an example of why a ‘parts list’ approach to the brain often leads to a dead end. To move the needle on therapy, we need to stop treating biophysics as an optional ‘fuzzy area’ and start using it as the foundational bedrock if we are ever going to make sense of the brain and why certain parts of it falter faster than others.
To expand on that, below are 6 tenets that point to why understanding biophysics is necessary to understanding the brain. While good neuroscience programs already require training in biophysics, the following need to be stressed early and often as they are foundational to understanding the brain and how it works. From these axioms the rest of our understanding can be built…
1. The Brain is an Electrical Circuit
The most fundamental argument is that neurons aren’t just cells; they are biological batteries and understanding how a thought moves requires the physics of electromagnetism.
- The Argument: Action potentials (nerve impulses) rely on voltage gradients and resistance.
- Physics Concept: Ohm’s Law and capacitance. Without the physics of cable theory, we couldn’t model how signals decay as they travel across long axons.
- Implication: Rethinking Disease as a “Circuit Failure” (Ohm’s Law & Capacitance). If the brain is fundamentally an electrical circuit, then neurodegenerative diseases like Parkinson’s may not just be about “missing” cells, but about signal decay and resistance. This implies that therapeutic interventions could move beyond replacing cells to “tuning” the circuitry to restore the proper voltage gradients and resistance required for healthy neural impulses. Rather than counting protein clumps we should be figuring out better methods for measuring voltage gradient decay in living systems.
2. Information Theory and Entropy
Neuroscience often asks what the brain is doing, but physics asks how efficiently it’s doing it.
- The Argument: The brain is an information-processing machine that must fight against noise and entropy.
- Physics Concept: Shannon Entropy. By applying information theory, we can measure the maximum capacity of a neural pathway and understand how the brain compresses data to save energy.
- Implication: Entropy as a Diagnostic. Clinical trials too often fail because we use clinical scales (like UPDRS). A physics-based measure (signal-to-noise ratio in motor circuits) would be more objective and more sensitive to early decay. Frame entropy as the Friction that wears down the Monster Truck’s engine regardless of how “well-built” (genetically normal) it is.
3. Statistical Mechanics and “Emergence”
You cannot understand a forest by looking at a single leaf; similarly, you can’t understand any brain process by looking at any one neuron.
- The Argument: The brain exhibits “emergent properties”—behaviors that arise only when billions of parts interact.
- Physics Concept: Phase transitions. Just as water turns to ice, the brain can shift between states (like wakefulness and sleep or anesthesia). Statistical mechanics provides the math to describe these collective behaviors.
- Implication: Preventing Systemic Collapse. Biologists too often treat aggregation as a “clump of trash,” while a biophysicist treats it as a predictable state change in a complex system. The difference is the ability to model the tipping point before it happens, rather than just cleaning up the mess afterward.
4. Energy Constraints (Thermodynamics)
The brain is an energy glutton, consuming about 20% of your body’s calories despite being only 2% of its weight.
- The Argument: The architecture of the brain is dictated by the need to minimize energy expenditure while maximizing speed.
- Physics Concept: The Laws of Thermodynamics. Evolution has “optimized” the brain’s wiring diagram to be a “small-world network,” a concept rooted in network physics.
- Implication: Thermodynamic Maintenance. The cell is not a static map, it is a dissipative structure. It only stays “alive” by burning energy to stay away from equilibrium (death). PD is a collapse of this flow. If the mitochondria can’t maintain the voltage gradient, the physics of the cell fails before the biology even notices.
5. The Scaling Problem
Why aren’t brains 10 times larger? Physics sets the limits.
- The Argument: As brains get bigger, the “wires” (axons) must get thicker to maintain speed, which takes up more space and requires more power.
- Physics Concept: Allometric scaling. There is a physical “ceiling” to intelligence based on heat dissipation and signal delay.
- Implication: Respect Physical Ceilings. No longer can we study the brain by isolating individual neurons or genes. The “Rules of the Road” apply from the molecule to the movement. If the protein folding is “chaotic,” the gait will eventually become “chaotic.” Physics allows us to link the micro to the macro using the same mathematical language.
6. Genetics ≠ Proteins
While genes provide the static digital code of an organism, proteins represent the dynamic physical execution of that code, subject to a wide variety of environmental and post-translational modifications that a genome alone cannot predict.
- The Argument: All the sequencing and genetic studies will not lead to a better understanding of neurodegenerative diseases nor therapies for them.
- Physics Concept: The Second Law of Thermodynamics and Gibbs Free Energy. Protein biophysics becomes critical to understanding how proteins are formed and how they interact with their environments. While there isn’t one single “Law of Protein,” the behavior of these macromolecules is dictated by the drive toward the lowest energy state.
- Implication: To quote Prof. John Hardy (one of the pioneers of neurogenetics): “I don’t think there is much left that needs to be explained about Alzheimer’s beyond the genetics.”*** This has become the dominant position in the field, that after Rosalind Franklin, Crick, Watson, and Wilkins discovered the structure of DNA using physical methods, there is no more physics needed to understand biology, all you need is more sequencing. The result was that we have mountains of sequencing data, with few meaningful models to interpret from or therapeutics based on that data that actually helps people live better lives.
To sum up, biology gives us a list of parts, physics gives us the manual, only when we combine our understanding of both disciplines are we able to interpret how those parts are used and why some of them break down over time. Without physics, we are just looking at a very complex “meat computer” without any understanding of the code or the electricity running through it. We must pivot from genomic cataloging to dynamic systems engineering. The next generation of diagnostics shouldn’t look for proteins; they should measure the bit-rate of neural circuits and the voltage gradients of mitochondria. Then we can start to frame diseases like Parkinson’s as not just clumps of this-or-that protein but as our brain’s inability to win the war against thermodynamic chaos.
* The Monstertruck analogy used above does not really convey the difference. Quoting Matt “The odds ratio of SNCA multiplications is >10 and that’s a MASSIVE effect, the odds ratio for Rs365165 (a GWAS SNP in SNCA best associated with Parkinson’s disease is 1.4…and is hardly greater than 1.0 (no effect). Although both genetic variants confer risk in the SNCA gene, their effect sizes are vastly different.”
**public database souces – Braineac (Ramasamy et al., 2014.Nat Neurosci. 17(10):1418–1428) and GTEx Portal (https://gtexportal.org/home/)
*** For more on that quote, and to better understand the argument made, I’d highly recommend visiting Kariem’s post on it here.