I have been thinking about protons a lot lately. Probably not too unusual for an old science guy. After all we used them in chemistry and physics. I have been doing some reading about stellar evolution lately and how the elements were formed. In that reading I came across the fact that a proton has a life of 10^32 years. Some estimates say 10^34 years. And proton decay doesn’t happen slowly over time. At some point it is just instantaneous. A proton is a composite particle rather than an elementary particle – composed of three valence quarks resulting in a net positive charge. But in chemistry and biochemistry they are generally written as a simple H+. When protons decay – it happens instantaneously to a positron and a pion. The positron is antimatter so it collides with an electron and is annihilated and gives off gamma photons. The pion explodes doing the same. The proton is converted to energy. This process is so rare it has never been directly observed although there is a massive experiment in progress to see if it can be done.
Practically all the protons in the universe today, were made
during the Big Bang about 13.8 billion years ago. Some protons are made in the universe today
but it is a very small process compared with the original source. That means
that all of the protons in my body (and yours) are recycled and will be for the
next 10^23 years. That’s nearly a
trillion trillion (10^24) years. The
various estimates for the death of the universe range from tens of billions of
years to 10^100 years. And of course,
life on Earth and Earth as a habitable planet is much shorter.
All of these thoughts about protons brings to mind Carl
Sagan’s various quotes about how we are all made of stardust. His thoughts go far beyond protons to every
element in our bodies and how they were synthesized in stars and temporarily
borrowed by us. Each one of us is an aggregate of this star dust maintained by
energy input and localization. Part of Sagan’s intent was to point out how this
can be reassuring and spiritual. I do
find it that way.
There is something else about protons. The verbal
description of protons may vary slightly between disciplines (physics, particle
physics, chemistry, biochemistry) but everyone is in agreement that protons
exist and can see the logic of all of the notations and definitions. Everyone agrees that there is one type of
proton and it will be around forever.
A lot of people will say species are qualitatively different
from disease and protons are qualitatively different from organisms and
diseases. Without using any
philosophical words – species, protons, and diseases have different rules of
existence, boundaries, and causal mechanisms.
To cite one example – no biological organism or disease originated in
the Big Bang and is expected to last forever. The rules of existence are much
different for a proton. It also does not adapt, process information, or experience
or feel anything. An organism has
emergent rather than summed properties, is subject to evolutionary pressure,
complex organization, and a finite lifespan. At the basic level the proton is a
thing and an organism is a process over time – a qualitative difference.
The comparison of species to disease shows that species are
generally individuals and diseases are abstract classes. The individuals are organized in an
evolutionary or phylogenic classification and diseases are organized as a disruption
of normal physiological or mental function in an individual. The species move through time and the disease
happens to an individual at one point in time. As the appreciation of disease
complexity has increased over time there is now an understanding that the lines
that blur the distinction between species can also occur with diseases.
That obviously is not the case in biology or medicine. From
taxonomy, we can find rare cases of organisms that are the only match to their
phenotype – there is one description of a single genus and species. We can also find identical phenotypes that
can be separated into genus and species only by genetic subtyping. We can find
monogenic diseases that produce several phenotypes and polygenic diseases that
produce many more. In other words – the match between description and consensus-based
reality is far from perfect.
I attempted to capture this phenomenon in the diagram at the
top of this post. I eschew the idea of a
spectrum or continuum, the categories are presented here just to
give estimates of possible numbers of subtypes in each category and the
associated uncertainty. Some may take issue with including speciation with
diseases. In that case I will defer to
Linnaeus who classified both and invented the predominant classification nomenclature
still used in taxonomy.
To be clear – this graphic is literally an apple to oranges
comparison. The first issue is the
physical wave-particle compared to biological entities. The second is the different levels within
biology. Species occur at the population
level over time defined by properties such as reproductive isolation,
morphological distinguishability, and monophyly. The boundaries between species reflect
evolutionary divergence rather than trait variation.
To cite few examples moving from left to right. The proton
is off the chart because there is no discrepancy between description and there
is only one agreed upon type. The first
three organisms (Gingko biloba, Balaeniceps rex, and Thermus aquaticus) all
have just one species through evolutionary mechanisms. I did not equate them to a proton because
being trained in biology (and without looking it up) – I am sure there are
various descriptions but common recognition they are the same species.
Things get interesting with the Bornean Fanged Frog. In this case all of the frogs look exactly
alike and can only be separated into 18 different species by molecular genetics
(1). The authors specific quote “single
species has been split into 18 genetically divergent yet morphologically
indistinguishable species.” This is the
definition of a cryptic species but they also point out controversy
about that definition and say that cryptic species are common in the Tree of
Life and understanding them is critical to understanding biodiversity.
Moving on to monogenic diseases, most of which can be
identified by molecular tests there are also limited number of phenotypes that
do not add much to classification but can be important in terms of treatment.
Infection and toxic agents are generally thought of as being defined by the
agent involved but there are a number of possible phenotypes based on host
susceptibility, organ system involved, disease state, lethality of the
infection agent or toxin, and in the case of the latter – toxidromes.
That brings me to major depression – one of the most
maligned diagnostic criteria in the DSM.
When I read the critiques, it seems like some people believe there will
be a magical verbal description of depression and all of our worries will be
over. The sun will shine and we will
never have to worry about actually treating the vagaries of depression on a clinical
basis. The new pure description will be perfect enough to lead to an
improvement in biological research and therapeutics. The other more insidious part of that
criticism is “I know more about depression than anybody and this is how it
should be classified and diagnosed.”
I don’t buy that criticism.
And here is why – I have seen tens of thousands of people with severe
depression and bipolar disorder successfully treated during the eras of the
DSM-III, DSM-IV, and DSM-5 and have worked with the thoughtful experts involved. I don’t think for a second that it matters
what was in those manuals or what turns up in the DSM-6. Assessing and treating depression and
differentiating it from other conditions doesn’t depend on what is in the DSM
or the ICD. It depends on what is in the
mind of the psychiatrist, how that mind was trained, and what that mind
experienced.
As far as the classification goes – I can show the table of
contents of Kraepelin’s text Clinical Psychiatry (2) to any practicing
psychiatrist today and they will recognize what he is talking about over a
hundred years ago. It seems that we have
to deny the validity of previous observations or label them as “pragmatic” but
otherwise meaningless. The newer hierarchical or network-based schemes don’t
mention the circularity of being based on descriptions pulled directly out of
the DSM and all previous observations.
The diagram shows that based on the DSM criteria there are
227 possible phenotypes of varying frequency and a recent study showed that
only 170 were observed in a large clinical sample. Genomic studies often use a compromised
phenotype by using the PHQ-9 or PHQ-2.
Nobody ever suggests that is a “practical” research compromise when you
are analyzing the genomes of many more people than several psychiatrists would
see in their lifetime. But that is one
reason some people think we need better criteria. How will better criteria be
useful in rapidly characterizing 100,000 people for a genomics study? Let me go out on a limb here and say there
will be no better verbal or written criteria.
There is a limit of what you can classify with just words – especially
in biology.
The proof is evident in the next three categories. Everyone can recognize a domestic dog. There
is tremendous phenotypic diversity in dogs based on morphology and
behavior. And they are all the same
genus and species. Atopic dermatitis or eczema is one of the most common
dermatological conditions and based on IgE status, age at onset, course,
endotype, molecular endotype, chronicity, fillagrin mutation status, and
severity there are 6,144 combinations although there is clinical overlap and
there has been no clinical investigation into how many of those variants exist.
From a morphological standpoint - many different rashes from eczema can exist
on the same person at the same time and specialists in dermatology are the best
people to diagnose that. The same
analysis can be done for systemic lupus erythematosus (SLE) using formal
criteria and that produces 27,648 combinations of signs, symptoms, and lab
findings.
There is a range to the limits of verbal classification in
biology and medicine. In the case of
cryptic species, we have a phenotype that presents very little perceptual or
verbal information for classification and that classification depends on
molecular biology. In some cases, there
is a one-to-one mapping of classification onto species. That rarely if ever works in medicine and
examples abound. I would not expect it to happen at high rates in biological
organisms with stochastic processes, genetic mechanisms like incomplete
penetrance, variable expressivity, polygenic modification of monogenic risk,
epistasis, pleiotropy, allelic heterogeneity, epigenetic variability, and
compound inheritance all increasing the gap between genotypes and expected
phenotypes. Approximate classifications
are not a deterrent to science or clinical practice even though that is a common
critical opinion.
Stay tuned for an even deeper dive into biological
classification of diseases based on some of these concepts.
George Dawson, MD, DFAPA
References:
1: Kin Onn Chan,
Dario N Neokleous, Shahrul Anuar, Rafe M Brown, Carl R Hutter, Indraneil Das,
Stefan T Hertwig, A Genomic Perspective on Cryptic Species Reveals Complex
Evolutionary Dynamics in the Gray Zone of the Speciation Continuum, Systematic
Biology, 2026;, syag001, https://doi.org/10.1093/sysbio/syag001
(open access).
2: Kraepelin E. Clinical Psychiatry. The MacMillan Company/Norwood Press,
Norwood,MA 1902, 1907. 628 p.
3: Kendler KS. The
Phenomenology of Major Depression and the Representativeness and Nature of DSM
Criteria. Am J Psychiatry. 2016 Aug 1;173(8):771-80. doi:
10.1176/appi.ajp.2016.15121509. Epub 2016 May 3. PMID: 27138588.
4: Zachar P, Kendler
KS. The Philosophy of Nosology. Annu Rev Clin Psychol. 2017 May 8;13:49-71.
doi: 10.1146/annurev-clinpsy-032816-045020. PMID: 28482691.
Graphics Credit:
An original from me - generated with MS Visio.
