LOW-FIDELITY BIOLOGY
The Missing Timing Layer in Development and Disease
“The question is not whether one exposure causes one diagnosis. The question is whether developing cells can lose the timing fidelity required to execute their instructions.”
Adapted from the discussion paper Bioelectric Fidelity in Early Human Development. https://www.rfsafe.com/wp-content/uploads/2026/08/Bioelectric_Fidelity_Early_Human_Development.pdf
I also need to explain why electromagnetic fields entered this question. I did not discover a recent organoid paper or the CYB5B study and then work backward until EMF fit the story. The concern came first. The molecular vocabulary came later.
Why Electromagnetic Fields Were Never A Guess
In 1995, the developmental-EMF literature was sparse, disputed, and largely invisible outside a narrow research community. Yet chick-embryo experiments had already reported that weak time-varying fields could alter early development and that pulse shape, frequency, and exposure window mattered. Other laboratories reported null results. The field was unresolved, but the question was real.
Then, in 1997, J. M. Farrell and colleagues published five years of experiments involving more than 2,500 White Leghorn embryos. Four campaigns used weak pulsed magnetic fields and one used a 60-hertz sinusoidal field. Four of the five campaigns showed statistically significant increases in abnormalities, ranging from twofold to sevenfold. Pooled pulsed exposure approximately doubled the abnormality rate, while the sinusoidal exposure approximately tripled it.
A companion study supplied an even clearer timing signal. A 4-microtesla, 60-hertz field increased the growth-related enzyme ornithine decarboxylase during gastrulation, then reduced it during neurulation. In a later morphology assessment, exposed embryos had about three times the control abnormality rate, with the vast majority of malformations involving the neural tube.
The same field shifted the same growth-related enzyme in opposite directions at different developmental stages. The biological meaning depended on when the field met the embryo. That was a developmental-timing finding before researchers had today’s vocabulary of calcium-waveform fidelity, chromatin readers, or mitochondrial transducers.
That research line also reported that temporally incoherent magnetic noise could suppress field-induced embryo abnormalities or ODC changes. Whatever final mechanism explains those experiments, temporal coherence itself had become a biological variable.
RF Safe was founded in 1998. The chronology matters. Personal loss raised the question. Embryo research made it experimentally concrete. RF Safe began. The following decades supplied more of the missing pathway.
Not every experiment was positive, and the Farrell series itself contained one nonsignificant campaign. But variability does not erase demonstrated biological capability. It may expose the controlling variables: waveform, coherence, developmental window, genetic background, baseline physiology, orientation, and exposure geometry.
A preponderance of evidence does not mean unanimity. The durable finding is not that every field always produces an effect. It is that weak, time-varying electromagnetic fields can alter biological signaling and development under defined conditions; temporal structure can matter; and later research has made the candidate pathway increasingly specific.
The central question is therefore not whether one exposure causes one diagnosis. It is whether developing cells can lose the timing fidelity required to execute their instructions.
That is the heart of the low-fidelity biology hypothesis.
Modern medicine is extraordinarily good at measuring genes, proteins, hormones, nutrients, toxins, receptors, and metabolites. But life depends just as much on when those parts act, where they act, and whether millions of events remain coordinated.
Biochemistry is the hardware layer. Timing is the execution layer.
A perfect instruction delivered at the wrong moment can become the wrong instruction. A growth signal that lasts too long can become overgrowth. A differentiation signal that arrives too late can leave progenitor cells in the wrong state. A contraction pulse that loses coordination with neighboring cells can alter tissue shape. A calcium signal with the correct average concentration but the wrong frequency, phase, or location can activate a different gene program.
Bioelectric fidelity is the accuracy with which cells generate, transmit, and decode voltage-dependent and ion-dependent signals. Low-fidelity biology is the degradation of that process. The signal becomes noisier, mistimed, less coherent, misplaced, or improperly coupled to the machinery meant to read it.
This is not another name for autism, cancer, autoimmunity, or metabolic disease. It is an upstream operating condition in which biology has less reserve and becomes more vulnerable to additional stress.
That is what I mean by a meta-disease state: an error-prone condition that can exist before a disease has a name. The energy spent buffering, correcting, and repairing that noise is what I call entropic waste.
Low fidelity changes probability; it does not determine destiny.
The Cell Does Not Read Calcium As A Bucket
To understand this idea, we have to stop thinking about calcium as if the only question were whether there is more of it or less of it.
A cell reads calcium as a waveform.
Imagine trying to understand the ocean by measuring only its average depth. That number tells you almost nothing about the waves. It does not tell you their frequency, height, direction, spacing, phase, or force. Two oceans could have the same average depth while one is calm and coherent and the other is chaotic and destructive.
Calcium works the same way. Cells decode the frequency of calcium oscillations, the height of each peak, the duration of each burst, the time between bursts, the location where calcium enters, the order in which compartments respond, the degree of synchronization between neighboring cells, and the speed with which the signal returns to baseline.
Classic experiments showed that calcium-oscillation frequency can change gene-expression efficiency and specificity. CaMKII can act as a frequency decoder, and calcium entering through a particular channel can signal to the nucleus differently from a global rise elsewhere. In the developing brain, calcium waves propagate through radial glia and influence proliferation.
So the scientifically relevant variable is not simply total calcium. It is calcium-code fidelity.
A developmental cell may receive nearly the same average calcium while the information contained in the waveform has changed. There may be extra bursts, missing bursts, altered phase, baseline drift, abnormal spatial propagation, or delayed recovery. Bulk assays can miss all of this.
An information channel can fail through timing noise before it fails through energy overload.
This is the gap in a safety system focused mainly on tissue heating. Heating and absorbed power are essential, but if living systems also respond to timing, average power cannot describe every relevant exposure feature.
Development Is Built From Critical Windows
Nowhere is timing more important than in embryonic development.
Development is a sequence of tightly ordered decisions. Cells divide, stop dividing, polarize, migrate, adhere, contract, differentiate, and communicate with adjacent tissues. Each step has a window, and the same signal can have different effects at different stages.
Neural-tube closure is one of the clearest examples.
During neurulation, cells do not simply squeeze themselves into shape with one continuous force. Live-imaging studies have shown that individual calcium flashes trigger pulsed contractions of the actin machinery that drives apical constriction. The timing and distribution of those pulses help bend and close the developing tissue.
Other experiments have shown that T-type calcium channels are required for normal neural-tube closure in animal models. Disrupting those channels reduced calcium transients and caused failures of anterior closure. Separate work found that different classes of intracellular calcium dynamics contribute differently to closure, and that disruption of the calcium pump SPCA1 alters cytoskeletal organization and neural-tube morphogenesis.
These studies do not show that electromagnetic exposure causes neural-tube defects. They establish that closure depends on organized calcium events, channel function, calcium homeostasis, and correctly timed contraction pulses.
A developing system can be pushed off course one mistimed calcium burst at a time – not because calcium is poison, but because timing is instruction.
That gives us a real experimental target. Do not begin by counting malformed embryos and then guessing why. Record the calcium flashes first. Record the contraction pulses. Measure their frequency, amplitude, phase, distribution, and recovery. Determine whether an exposure changes the signal before the anatomy changes. Then replay or restore the waveform and see whether the developmental outcome follows.
That is how a hypothesis becomes a mechanism.
The S4-Mito-Spin Framework
The S4-Mito-Spin framework is my attempt to describe the cellular circuitry through which an electromagnetic environment could influence biological timing.
It is not one magic receptor. It is an integrated architecture with three interacting layers.
The first layer is S4.
Many voltage-gated ion channels contain charged S4 voltage-sensing structures. These are nanoscale components that move in response to changes in membrane voltage and help determine whether a channel opens or closes. When a calcium channel opens, it does not merely add calcium to a bucket. It creates a localized microdomain with a particular timing relationship to calmodulin, kinases, mitochondria, the endoplasmic reticulum, and the nucleus.
Published ion forced-oscillation models propose that polarized anthropogenic fields may perturb membrane-associated ions and voltage-sensitive channel machinery. The model remains debated, and ordinary wireless exposure has not been proven to force S4 gates open. But S4 is a specific candidate that can be tested with channel mutants, knockouts, rescue, and voltage clamp.
The second layer is Mito – the mitochondria.
Mitochondria are not just power plants. They are timing hubs. They buffer calcium, shape calcium transients, generate ATP, regulate redox state, communicate with the endoplasmic reticulum, and participate in decisions about growth, repair, inflammation, and cell death.
A membrane disturbance can be amplified by mitochondrial calcium uptake and redox signaling, while altered electron transfer can reshape the waveform returned to the cell. Voltage affects calcium; calcium affects mitochondria; mitochondrial redox state affects channels, enzymes, and gene regulation.
The third layer is Spin.
Spin refers to the quantum-chemical behavior of electrons in radical pairs, flavins, transition metals, and heme-containing proteins. This is the layer where weak electromagnetic inputs may influence reaction probabilities without acting like a source of bulk heat.
For years, Spin was the hardest layer to connect to named cellular hardware. Then, in 2026, Kim and colleagues used a CRISPR screen in an electromagnetic-field-inducible gene-switch study and identified cytochrome b5 type B, or CYB5B, as an essential mediator likely acting as a sensor in their engineered system.
CYB5B is an outer-mitochondrial-membrane heme protein involved in electron transfer. Its heme group makes it a concrete target for testing redox and spin-sensitive mechanisms. But we must be precise: the Kim study did not prove that Wi-Fi spin-biases CYB5B, and it did not establish the exact physical coupling mechanism. That requires spectroscopy, field-orientation testing, electron-transfer measurements, and rescue with heme-binding or redox mutants.
What the study did show is still extremely important. The engineered response required CYB5B, and activation depended on rhythmic calcium oscillations rather than generic calcium influx.
In other words, the cell did not respond merely because calcium increased. It responded because a particular calcium rhythm was produced and decoded.
That finding lands directly on the central prediction of low-fidelity biology: timing is the biological payload.
S4, Mito, and Spin can therefore converge on the same variable. S4 represents a candidate membrane-gating route. Mito represents the redox-calcium amplifier and coordinator. Spin represents field-sensitive electron and heme chemistry, with CYB5B now providing a named experimental candidate.
If one layer perturbs calcium entry and another perturbs mitochondrial rhythm or redox timing, the result may not be a dramatic calcium flood. It may be something subtler and more consequential: a waveform that still looks like calcium signaling, but carries degraded information.
That is bioelectric dissonance.
Why The Temporal Structure Of Wireless Signals Matters
A wireless signal is not adequately described by naming only its carrier frequency.
A 2.4-gigahertz Wi-Fi or Bluetooth carrier also contains packet timing, bursts, duty cycles, beacon intervals, modulation patterns, and changes in amplitude over time. A common Wi-Fi beacon pattern can recur roughly ten times per second. Legacy GSM contains a well-known 217-hertz time structure. Other systems use different and often adaptive patterns.
A low-frequency envelope on a gigahertz carrier is not physically identical to a standalone low-frequency magnetic field. The carrier, electric-to-magnetic ratio, polarization, geometry, induced currents, and tissue coupling differ. We should not collapse them into the same exposure.
But it is equally incomplete to point only to the gigahertz carrier and ignore the temporal structure carrying the information.
Biological systems are nonlinear. They contain thresholds, rectifying membranes, voltage sensors, coupled oscillators, and redox reactions. The correct question is whether any cellular transducer detects the carrier, the envelope, their interaction, or neither.
That question is testable.
Compare continuous-wave and modulated carriers at the same absorbed power, a pure low-frequency field reproducing the envelope frequency, and a matched thermal control. Record the full waveform, internal dose, orientation, temperature, and biological response.
If modulation changes calcium timing while continuous wave does not, average power is not the whole story. If both exposures are indistinguishable whenever absorbed power and temperature are matched, the timing hypothesis is weakened.
This is why the framework is scientific. It can lose.
Three Modern Studies That Form A Testable Bridge
The early embryo studies established a developmental warning and showed that temporal structure could matter. Modern molecular and organoid research now provides three complementary anchors that turn that warning into a directly testable pathway.
The first anchor is the 2025 study by Cakir and colleagues on human cortical organoids.
Cortical organoids are three-dimensional stem-cell models of early cortical development. They do not contain a placenta, skull, full immune system, normal vasculature, maternal metabolism, or complete fetal physiology, but they allow researchers to observe early human neural progenitor decisions.
Cakir and colleagues exposed developing cortical organoids to a defined 2.4-gigahertz Bluetooth-like source. The reported transmitter output was approximately 2.4 milliwatts, with a reported maximum measured external power density around 2.5 milliwatts per square meter. Those external numbers are not a substitute for mapping absorbed dose inside each organoid, and independent replication must improve the dosimetry.
Within that model, however, the reported biological findings were significant. Radiofrequency exposure altered radial-glia differentiation and kept more cells in a stem-like state for longer. It changed transcriptional and chromatin-accessibility programs. It increased expression of autism-associated genes and endogenous retroelements. Neurons that formed under exposure showed altered morphology, synaptic features, electrical behavior, and spontaneous calcium activity.
Most importantly, low-dose inhibitors of BET-family proteins rescued multiple radiofrequency-associated effects, including developmental, transcriptional, and neuronal phenotypes.
BET proteins are epigenetic readers. They bind acetylated lysines on histones and other proteins and help recruit the transcriptional machinery that determines which genes are actively read. The rescue does not mean BET is the radiofrequency sensor. It means BET activity is a functional downstream node in the altered developmental state.
That distinction is crucial.
The second anchor is the Kim CYB5B study.
Kim and colleagues used a defined 60-hertz, 2.0-millitesla electromagnetic field in an engineered gene-switch system. This was not Wi-Fi or Bluetooth. But the CRISPR result identified CYB5B as necessary for the field-responsive pathway, and the calcium result showed that rhythmic oscillations – not generic calcium elevation – carried the signal into gene activation.
This gives us a candidate upstream transducer and a timing-dependent intermediate.
The third anchor is the 2024 work by Courchesne and colleagues.
That team generated 4,910 brain cortical organoids from deeply characterized autistic toddlers and controls. The autism-derived organoids were larger and grew faster on average. The largest organoids were associated with more severe later social symptoms and accelerated neurogenesis. The study also found reduced NDEL1 activity and expression associated with organoid growth characteristics.
NDEL1 is involved in centrosomal function, mitosis, dynein regulation, neuronal migration, neurite growth, and morphogenesis. Its activity is regulated by phosphorylation. That makes it a reasonable candidate within a timing-sensitive kinase network.
But the Courchesne study did not expose organoids to radiofrequency fields, and it did not show that NDEL1 suppression is the single cause of profound autism. NDEL1 belongs in the framework as a developmental-control branch that must be measured, not as a link we are allowed to assume.
Now place the three studies side by side.
Cakir shows a radiofrequency-associated, BET-dependent developmental phenotype in human cortical organoids.
Kim shows a CYB5B-dependent electromagnetic response carried by rhythmic calcium dynamics in a separate engineered system.
Courchesne shows NDEL1-associated growth and neurogenesis differences in autism-derived cortical organoids.
No single experiment connects all three. That missing connection is not a reason to ignore the pathway. It is the experiment the evidence now tells us to perform.
From Calcium Timing To Chromatin And Cell Fate
The proposed bridge is biologically specific.
A time-structured electromagnetic input interacts with a susceptible transducer – potentially CYB5B, a voltage-gated channel, or both. The resulting change is not defined simply as more calcium. It is a change in calcium frequency, phase, localization, burst structure, or recovery.
Calcium-dependent enzymes then decode that waveform. CaMKII is sensitive to frequency. Calcineurin and NFAT integrate repeated calcium events over time. PKC and other kinases respond according to their own activation and deactivation kinetics. These pathways regulate transcription factors and also influence histone acetyltransferases and histone deacetylases.
That creates the translation layer between calcium timing and chromatin state.
BET proteins then read the acetylation landscape. BRD2, BRD3, and BRD4 do not need to be directly struck by radiofrequency energy. They may simply do their normal job on a chromatin landscape that was written at the wrong time.
The downstream machinery can function as designed and still produce the wrong developmental output if the upstream timing code is corrupted.
BET may faithfully read a mistimed signal.
In parallel, altered kinase timing may affect NDEL1 expression, phosphorylation, location, or activity. The BET branch and the NDEL1 branch could then converge on radial-glia self-renewal, differentiation, mitosis, migration, and cortical organization.
This is not an unbroken causal chain that has already been demonstrated. It is a coherent adverse-outcome pathway containing direct findings, established biological bridges, and precisely defined missing links.
The Meta-Disease State
This brings us back to low-fidelity biology as a meta-disease state.
This framework is not a theory of moral decline. Autism and neurodivergence do not measure human worth, empathy, identity, or social value. The scientific focus is preventable impairment, high-support-needs outcomes, and biological timing – not ranking people.
A meta-disease state does not mean that one electromagnetic exposure directly causes every modern illness. Shared machinery does not establish a shared cause. Cancer, autoimmunity, metabolic disease, infertility, and neurodevelopmental disorders each require their own disease-specific pathway and evidence.
The narrower proposition is that many tissues rely on the same upstream control architecture: voltage sensing, calcium timing, mitochondrial redox signaling, chromatin regulation, and coordinated gene expression. If the fidelity of that architecture is reduced, different tissues may fail in different ways.
In a developing cortex, the consequence may be a mistimed progenitor decision or altered circuit maturation. In a closing neural tube, it may be disordered contraction pulses. In an immune cell, it may be altered activation thresholds or tolerance signaling. In a metabolically active tissue, it may be reduced synchronization between fuel demand, insulin signaling, and mitochondrial output. In a proliferating cell, it may increase the burden on cell-cycle checkpoints and repair systems.
These are not established outcomes of ordinary wireless exposure. They explain why an upstream-fidelity hypothesis predicts diverse, tissue-specific vulnerability rather than one exposure producing one uniform disease.
The model is explicitly multifactorial. Genes, folate and other nutrients, infection, inflammation, oxygenation, circadian timing, air pollution, medications, metabolism, endocrine state, chemical exposures, and physical environmental inputs can all matter.
A developing system may compensate for one mild disturbance. It may fail when several converge in the same narrow window. The key concepts are reserve, threshold, interaction, and susceptibility. True transgenerational persistence is a separate question and would require effects to continue into an unexposed generation; it is not established here.
This is why low fidelity is not destiny. It is a change in the probability landscape.
It can also explain variable responders. Two people can receive a similar external exposure while experiencing different internal doses, channel expression, mitochondrial states, genetic buffering, developmental timing, and compensatory capacity. A null average across a mixed population does not prove that no susceptible subgroup exists. At the same time, a susceptible subgroup cannot be assumed; it must be found with biomarkers and prospective measurement.
The Population Question
The identified prevalence of autism has risen dramatically. CDC surveillance for 2022 estimated approximately one in thirty-one eight-year-old children across the participating United States sites, with substantial variation among locations and identification practices.
The expansion of household wireless technology occurred across overlapping decades. That co-trend is not proof of causation. Changes in diagnostic criteria, awareness, access to services, ascertainment, parental age, survival, genetics, and other environmental conditions all contribute to the observed pattern.
But a co-trend is a legitimate reason to ask a mechanistic question, especially when laboratory studies reveal biologically specific points of contact.
The correct response is neither to use the timeline as proof nor to declare the question closed because the timeline is ecological.
A serious human study would measure personal maternal exposure before conception and throughout pregnancy, including distance, nighttime and occupational sources, waveform characteristics, and fetal dosimetry. It would also measure folate, genetics, infection, inflammation, medications, pollution, metabolism, and biomarkers connecting exposure to calcium or chromatin changes.
Autism should not be treated as one biological endpoint. The Courchesne organoid work itself points to different growth subtypes. A mechanism may apply strongly to one developmental subgroup and weakly or not at all to another.
Averaging everyone together can erase the very signal we need to find.
The Experiment That Must Now Be Done
The decisive experiment is straightforward in concept, even if technically demanding.
First, independently replicate the Cakir 2.4-gigahertz organoid findings with blinded exposure codes, matched shams, preregistered endpoints, complete dosimetry, and thermal controls.
Second, manipulate CYB5B. Use inducible knockdown, knockout if viable, wild-type rescue, and mutants affecting heme binding, electron transfer, membrane anchoring, and protein interactions. If CYB5B is necessary for the radiofrequency-associated calcium and BET phenotype, reducing it should change the response and wild-type rescue should restore it.
Third, test the S4 route rather than merely arguing about it. Use channel knockouts, voltage-sensor mutants, rescue experiments, and patch-clamp measurements during exposure.
Fourth, measure calcium as information. Record frequency, amplitude, phase, burst duration, propagation, spectral entropy, recovery, compartment coupling, and cell-to-cell synchronization. The most revealing result would be a condition in which average calcium remains similar while temporal organization changes and developmental output follows.
Fifth, map the hierarchy from calcium to chromatin. Measure kinase and phosphatase activity, histone acetylation, chromatin accessibility, BET occupancy, RNA polymerase pausing, and single-cell gene expression. Then use controlled calcium-waveform replay to determine whether the altered pattern is sufficient to reproduce the chromatin and cell-state changes.
Sixth, test whether BET is truly downstream. BET inhibition should rescue transcription and differentiation without necessarily correcting the initial CYB5B redox change or calcium waveform. Restoring the waveform should normalize BET behavior without directly blocking BET.
Seventh, measure NDEL1 directly – its expression, enzymatic activity, phosphorylation, localization, mitotic effects, and rescue potential.
Finally, move into human neuruloids, neural-tube organoids, and validated vertebrate systems. Record calcium flashes and actomyosin contraction pulses during exposure. Do not infer timing failure from anatomy after the fact. Watch the failure emerge, or fail to emerge, in real time.
The framework must also specify what would weaken it.
If independent laboratories cannot reproduce the organoid phenotype, confidence must fall. If correcting temperature, vibration, handling, or incubator artifacts removes the effect, the claim fails. If no calcium-waveform disturbance precedes the chromatin changes, the proposed direction is wrong. If CYB5B and channel manipulation do nothing, those candidates should be rejected. If continuous-wave and modulated exposure are identical whenever dose and temperature are matched, modulation specificity is weakened. If realistic fetal dosimetry falls far below any replicated active range, the human-developmental interpretation must be revised.
A negative result is progress. It prevents a legitimate concern from becoming an unfalsifiable story.
The 140-Year Mistake
When I speak of a 140-year mistake, I am not saying that electromagnetism or wireless communication should never have been developed.
I am saying that electromagnetic technology has been evaluated mainly as an energy-transfer problem, while living systems are also information-processing systems.
The correction is not to throw away dosimetry. It is to add the missing variables: waveform, modulation, timing, polarization, orientation, internal field distribution, developmental window, and biological signal fidelity.
We do not need to abandon the digital age. We need to engineer it with a better biological model.
That means wired and fiber backbones wherever practical. It means eliminating unnecessary transmitters, increasing distance, reducing duty cycle, and designing devices around children and pregnancy rather than treating every body as an interchangeable adult model. It means developing optical wireless systems such as Li-Fi with rigorous standards for wavelength, irradiance, flicker, modulation depth, eye and skin safety, and photosensitive populations.
Safety is an engineering specification, not a slogan.
A Clean Aether policy should be technology-neutral and evidence-driven. It should require exposure transparency, full waveform reporting, independent biological research, child-sensitive design, wired and optical compatibility, and continuous evaluation as the science develops.
The Conclusion
I am not asking science to declare that wireless radiation causes autism, neural-tube defects, or every chronic disease. I am asking science to recognize the continuity of the evidence and test the mechanism it now points toward.
In 1997, the warning appeared as embryo abnormalities and stage-specific changes in a growth-related enzyme. When I founded RF Safe in 1998, we did not yet have live calcium imaging, human cortical organoids, single-cell sequencing, chromatin mapping, CRISPR screens, or a named mitochondrial candidate such as CYB5B.
Today, we know that calcium waveforms carry developmental information and that neural-tube closure and cortical development depend on timed calcium events. We have a human cortical-organoid study reporting radiofrequency-associated, BET-dependent developmental changes. We have a CRISPR study identifying CYB5B as essential in an electromagnetic-field-responsive pathway driven by rhythmic calcium oscillations. And we have autism-derived organoid research connecting accelerated growth and neurogenesis with reduced NDEL1 activity and expression.
The complete chain is not proven. But the concern was not invented after the latest paper. The pathway has become more specific, measurable, and falsifiable.
The preponderance claim is one of capability and mechanistic plausibility, not universal causation: time-varying fields can alter biology under defined conditions; temporal structure can matter; development depends on timing; and we now possess the tools to determine whether realistic exposures degrade that timing in susceptible human developmental systems.
That is enough to make the next step an obligation, not an option. The science is not finished. The decisive experiment is overdue.
Angel Leigh’s life is not evidence of a cause. It is the reason I refuse to let this question be reduced to slogans. Her legacy, for me, is a demand to measure the live code of development, not merely the chemistry left behind after the code has run.
Can chronic, time-structured electromagnetic input reduce the fidelity of the calcium signals that developing cells use to make irreversible decisions?
We now have the tools to answer it.
We are not asking science to accept a conclusion. We are asking it to stop leaving the decisive experiment undone.
Research References
Not intended for narration. These sources support the empirical anchors and established biological bridges described in the script. Proposed links remain hypotheses until tested in a unified exposure model.
Historical developmental-field evidence
1. Delgado JMR, Leal J, Monteagudo JL, Gracia MG. Embryological changes induced by weak, extremely low frequency electromagnetic fields. Journal of Anatomy. 1982;134(Pt 3):533-551.
2. Ubeda A, Leal J, Trillo MA, Jimenez MA, Delgado JMR. Pulse shape of magnetic fields influences chick embryogenesis. Journal of Anatomy. 1983;137(Pt 3):513-536.
3. Maffeo S, Miller MW, Carstensen EL. Lack of effect of weak low frequency electromagnetic fields on chick embryogenesis. Journal of Anatomy. 1984;139(Pt 4):613-618.
4. Juutilainen J, Harri M, Saali K, Lahtinen T. Effects of 100-Hz magnetic fields with various waveforms on the development of chick embryos. Radiation and Environmental Biophysics. 1986;25(1):65-74. doi:10.1007/BF01209686.
5. Berman E, Chacon L, House D, et al. Development of chicken embryos in a pulsed magnetic field. Bioelectromagnetics. 1990;11(2):169-187. doi:10.1002/bem.2250110208.
6. Ubeda A, Trillo MA, Chacon L, Blanco MJ, Leal J. Chick embryo development can be irreversibly altered by early exposure to weak extremely-low-frequency magnetic fields. Bioelectromagnetics. 1994;15(5):385-398. doi:10.1002/bem.2250150503.
7. Litovitz TA, Montrose CJ, Doinov P, Brown KM, Barber M. Superimposing spatially coherent electromagnetic noise inhibits field-induced abnormalities in developing chick embryos. Bioelectromagnetics. 1994;15(2):105-113. doi:10.1002/bem.2250150203.
8. Farrell JM, Litovitz TL, Penafiel M, Montrose CJ, Doinov P, Barber M, Brown KM, Litovitz TA. The effect of pulsed and sinusoidal magnetic fields on the morphology of developing chick embryos. Bioelectromagnetics. 1997;18(6):431-438. doi:10.1002/(SICI)1521-186X(1997)18:6<431::AID-BEM5>3.0.CO;2-3.
9. Farrell JM, Barber M, Krause D, Litovitz TA. Effects of low frequency electromagnetic fields on the activity of ornithine decarboxylase in developing chicken embryos. Bioelectrochemistry and Bioenergetics. 1997;43(1):91-96. doi:10.1016/S0302-4598(96)05174-4.
10. Farrell JM, Barber M, Krause D, Litovitz TA. The superposition of a temporally incoherent magnetic field inhibits 60 Hz-induced changes in the ODC activity of developing chick embryos. Bioelectromagnetics. 1998;19(1):53-56. doi:10.1002/(SICI)1521-186X(1998)19:1<53::AID-BEM6>3.0.CO;2-3.
Core experimental anchors
11. Cakir B, Tanaka Y, Choe MS, et al. Radiofrequency regulates the BET-mediated pathways in radial glia differentiation in human cortical development. Cell Reports. 2025;44(10):116238. doi:10.1016/j.celrep.2025.116238.
12. Kim J, Hwang Y, Kim S, et al. Electromagnetic field-inducible in vivo gene switch for remote spatiotemporal control of gene expression. Cell. 2026;189(11):3465-3480.e23. doi:10.1016/j.cell.2026.03.029.
13. Kim J, Hwang Y, Kim S, et al. Erratum to: Electromagnetic field-inducible in vivo gene switch for remote spatiotemporal control of gene expression. Cell. 2026;189(14):4517-4518. doi:10.1016/j.cell.2026.06.020.
14. Courchesne E, Taluja V, Nazari S, et al. Embryonic origin of two ASD subtypes of social symptom severity: the larger the brain cortical organoid size, the more severe the social symptoms. Molecular Autism. 2024;15:22. doi:10.1186/s13229-024-00602-8.
Calcium-waveform coding and developmental timing
15. Dolmetsch RE, Xu K, Lewis RS. Calcium oscillations increase the efficiency and specificity of gene expression. Nature. 1998;392(6679):933-936. doi:10.1038/31960.
16. De Koninck P, Schulman H. Sensitivity of CaM kinase II to the frequency of Ca2+ oscillations. Science. 1998;279(5348):227-230. doi:10.1126/science.279.5348.227.
17. Dolmetsch RE, Pajvani U, Fife K, Spotts JM, Greenberg ME. Signaling to the nucleus by an L-type calcium channel-calmodulin complex through the MAP kinase pathway. Science. 2001;294(5541):333-339. doi:10.1126/science.1063395.
18. Weissman TA, Riquelme PA, Ivic L, et al. Calcium waves propagate through radial glial cells and modulate proliferation in the developing neocortex. Neuron. 2004;43(5):647-661. doi:10.1016/j.neuron.2004.08.015.
19. Christodoulou N, Skourides PA. Cell-autonomous Ca2+ flashes elicit pulsed contractions of an apical actin network to drive apical constriction during neural tube closure. Cell Reports. 2015;13(10):2189-2202. doi:10.1016/j.celrep.2015.11.017.
20. Abdul-Wajid S, Morales-Diaz H, Khairallah SM, Smith WC. T-type calcium channel regulation of neural tube closure and EphrinA/EPHA expression. Cell Reports. 2015;13(4):829-839. doi:10.1016/j.celrep.2015.09.035.
21. Suzuki M, Sato M, Koyama H, et al. Distinct intracellular Ca2+ dynamics regulate apical constriction and differentially contribute to neural tube closure. Development. 2017;144(7):1307-1316. doi:10.1242/dev.141952.
22. Brown JM, Garcia-Garcia MJ. Secretory pathway calcium ATPase 1 (SPCA1) controls mouse neural tube closure by regulating cytoskeletal dynamics. Development. 2018;145(19):dev170019. doi:10.1242/dev.170019.
Chromatin, BET proteins, and NDEL1
23. Filippakopoulos P, Qi J, Picaud S, et al. Selective inhibition of BET bromodomains. Nature. 2010;468(7327):1067-1073. doi:10.1038/nature09504.
24. Jang MK, Mochizuki K, Zhou M, Jeong HS, Brady JN, Ozato K. The bromodomain protein Brd4 is a positive regulatory component of P-TEFb and stimulates RNA polymerase II-dependent transcription. Molecular Cell. 2005;19(4):523-534. doi:10.1016/j.molcel.2005.06.027.
25. Backs J, Song K, Bezprozvannaya S, Chang S, Olson EN. CaM kinase II selectively signals to histone deacetylase 4 during cardiomyocyte hypertrophy. Journal of Clinical Investigation. 2006;116(7):1853-1864. doi:10.1172/JCI27438.
26. Woo Y, Kim SJ, Suh BK, et al. Sequential phosphorylation of NDEL1 by the DYRK2-GSK3beta complex is critical for neuronal morphogenesis. eLife. 2019;8:e50850. doi:10.7554/eLife.50850.
27. Mori D, Yano Y, Toyo-oka K, et al. NDEL1 phosphorylation by Aurora-A kinase is essential for centrosomal maturation, separation, and TACC3 recruitment. Molecular and Cellular Biology. 2007;27(1):352-367. doi:10.1128/MCB.00878-06.
CYB5B physiology and proposed electromagnetic transduction
28. Ma MY, Deng G, Zhu WZ, et al. Defects in CYB5A and CYB5B impact sterol-C4 oxidation in cholesterol biosynthesis and demonstrate regulatory roles of dimethyl sterols. Cell Reports. 2024;43(11):114912. doi:10.1016/j.celrep.2024.114912.
29. Jakobs HH, Mikula M, Havemeyer A, et al. The N-reductive system composed of mitochondrial amidoxime reducing component (mARC), cytochrome b5 (CYB5B) and cytochrome b5 reductase is regulated by fasting and high-fat diet in mice. PLoS ONE. 2014;9(8):e105371. doi:10.1371/journal.pone.0105371.
30. Struwe MA, Clement B. The mitochondrial amidoxime reducing component – from prodrug activation to metabolic regulation. Journal of Biological Chemistry. 2023;299(10):105306. doi:10.1016/j.jbc.2023.105306.
31. Panagopoulos DJ, Yakymenko I, De Iuliis GN, Chrousos GP. A comprehensive mechanism of biological and health effects of anthropogenic extremely low frequency and wireless communication electromagnetic fields. Frontiers in Public Health. 2025;13:1585441. doi:10.3389/fpubh.2025.1585441.
32. Panagopoulos DJ, Karabarbounis A, Yakymenko I, Chrousos GP. Human-made electromagnetic fields: ion forced-oscillation and voltage-gated ion channel dysfunction, oxidative stress and DNA damage. International Journal of Oncology. 2021;59(5):92. doi:10.3892/ijo.2021.5272.
Population and policy context
33. Shaw KA, Williams S, Patrick ME, et al. Prevalence and early identification of autism spectrum disorder among children aged 4 and 8 years – Autism and Developmental Disabilities Monitoring Network, 16 sites, United States, 2022. MMWR Surveillance Summaries. 2025;74(2):1-22.
34. Environmental Health Trust v. Federal Communications Commission, 9 F.4th 893 (D.C. Cir. 2021).
35. Greene NDE, Copp AJ. Neural tube defects. Annual Review of Neuroscience. 2014;37:221-242. doi:10.1146/annurev-neuro-062012-170354.
36. Skinner MK. What is an epigenetic transgenerational phenotype? F3 or F2. Reproductive Toxicology. 2008;25(1):2-6. doi:10.1016/j.reprotox.2007.09.001.
