Before it Begins: A Preventative Multi Layer Calcium Driven Framework for Alzheimer’s

Hi, my name is Sherdha Sharma, I am a grade 9 student from Edmonton, Alberta.

Transcript: Alzheimer's affects millions worldwide, yet most treatments only target symptoms after significant brain damage has already occurred. This raises a critical question: What if we could intervene earlier? I compared treatments such as Donepezil, Memantine, and amyloid therapies, as well as theories and analyzed genes involved in Alzheimer's to identify a plausible root cause. I found that current approaches largely mask downstream effects instead of the underlying biology. To investigate this, I developed deep neurogenics using Streamlit, a model that analyzes early biomarkers and simulates preclinical dysfunction. Through analysis and identifying gaps in research, I found calcium dysregulation to be a key factor. Based on this, I proposed Precision Epigenetic Rebalancing or PER, a theoretical gene regulation strategy using AAV9 to restore balance and target the origin of disease. This work highlights shifting Alzheimer's care from reactive treatment to early mechanism driven prevention.

Alzheimer’s disease (AD) is a progressive neurodegenerative disorder affecting millions worldwide and is typically diagnosed only after cognitive symptoms appear. However, evidence suggests that underlying molecular dysfunction begins years before, limiting the effectiveness of current treatments (Jack et al., 2018).

I was inspired to pursue this project due to my experiences with MTBIS, chronic illness, and watching family navigate serious health challenges. This led me to question why neurological conditions are often treated only after significant damage has occurred, and whether earlier biological changes could be identified. To investigate this, I first developed DeepNeuroGenX using Streamlit, building a computational model to simulate early biomarker patterns and neuronal instability. I then conducted structured literature analysis, comparing pathways such as BACE1 expression, calcium signaling, and synaptic dysfunction to identify shared mechanisms.

Through iterative research and model refinement, calcium dysregulation emerged as a central driver affecting synaptic function, gene expression, and mitochondrial stability, in particular, altered NMDA and ryanodine receptor activity. (Berridge, 2010). After examining treatment and researchs, I applied similar targeted intervention concepts to develop Precision Epigenetic Rebalancing (PER) as a novel, theoretical strategy aimed at restoring regulatory equilibrium prior to irreversible neurodegeneration through AAV9 mediated gene regulation.

By prioritizing early stabilization of underlying brain systems, this framework redefines Alzheimer’s intervention from reactive treatment to proactive prevention, with the goal of preserving neural function, slowing disease progression, and ultimately improving outcomes for individuals before symptoms appear.

To develop my solution, I used a multi stage research, systems analysis, and computational design process to build DeepNeuroGenX and Precision Epigenetic Rebalancing (PER). I began by analyzing major Alzheimer’s theories, including amyloid, tau, and calcium based models, using peer reviewed literature. While each explained part of the disease, they were often studied in isolation. By comparing findings, I identified a consistent pattern: disruptions in neuronal signaling, especially NMDA receptor activity, calcium regulation, and neural oscillations, were linked to downstream changes in gene expression, mitochondrial dysfunction, and synaptic instability. This suggested Alzheimer’s may result from multi layer system instability rather than a single pathway failure.

I then developed DeepNeuroGenX using Python and Streamlit as a predictive platform integrating imaging, genetic, and symptom data to generate personalized 10 year risk profiles and reports, designed to be accessible and free. This allowed me to explore how different combinations of biomarkers could influence disease progression and highlight variability between individuals.

Using these insights, I designed PER as a molecular level regulatory framework that modulates gene expression, including BACE1, through epigenetic mechanisms while stabilizing upstream signaling to prevent calcium overload and downstream damage such as mitochondrial dysfunction and synaptic loss. This approach was informed by examining how gene dysregulation contributes to amyloid imbalance and neuronal stress.

Rather than blocking pathways, PER rebalances interacting systems across molecular, cellular, and network levels. Finally, I validated the framework by ensuring consistency across studies and by comparing it to current treatments such as Donepezil, Memantine, and amyloid targeting therapies, which primarily act on isolated or late stage processes rather than early system instability.

The results of this project show that Alzheimer’s disease can be more effectively understood as a multi layer system instability rather than a single pathway disorder. By comparing findings across multiple peer reviewed studies, I identified consistent patterns linking disrupted neuronal signaling, gene expression, and large scale brain dysfunction.

As shown in Figure 1, increased NMDA receptor (NMDAR) activity and calcium dysregulation were repeatedly associated with elevated BACE1 expression, mitochondrial stress, and synaptic damage. This pathway model summarizes how early molecular disruptions can propagate across cellular and network levels, eventually leading to impaired neural oscillations and cognitive decline. The consistency of this pathway across independent studies supports its validity as a unified model of disease progression.

To apply these findings, I developed DeepNeuroGenX, a predictive platform that demonstrates how these biological patterns can be used for early detection. As illustrated in Figure 2, the system integrates multiple inputs, including imaging data (MRI, EEG), biomarker indicators, genetic information, and symptom patterns. These inputs are processed to generate a personalized risk profile and a projected disease progression over a 10 year period. The results vary between individuals, showing that different biomarker combinations lead to different predicted trajectories, supporting the need for personalized approaches.

Using this system level understanding, I evaluated the proposed Precision Epigenetic Rebalancing (PER) framework. As shown in Figure 3, PER targets the same pathway identified in Figure 1, but at the molecular level. It works by modulating gene expression through epigenetic mechanisms, including adjusting chromatin accessibility, regulating transcription rates, and influencing mRNA stability of key genes such as BACE1 and its antisense regulator. This allows controlled reduction of harmful protein production while preserving essential biological function.

In addition, PER stabilizes upstream signaling by moderating NMDAR activity and preventing calcium overload, which reduces downstream effects such as kinase dysregulation, mitochondrial dysfunction, and synaptic loss. Rather than blocking individual pathways, PER operates within a controlled range to rebalance interacting systems.

Overall, the results demonstrate that integrating biomarkers, signaling pathways, and gene regulation into a unified framework provides a more predictive and biologically consistent understanding of Alzheimer’s disease. This approach not only explains disease progression across multiple levels but also supports earlier, more personalized intervention strategies compared to current treatments.To further evaluate these results, the framework was compared to existing Alzheimer’s treatments, including Donepezil, Memantine, and amyloid targeting therapies. These treatments primarily act on single pathways or late stage effects, such as neurotransmitter levels or plaque removal, without addressing the upstream causes of disease progression. In contrast, PER targets multiple interacting systems, including neuronal signaling, gene expression, and cellular stability, allowing earlier and more comprehensive intervention. This comparison supports the conclusion that addressing system level instability may provide a more effective and sustained approach than current treatment strategies.

The results of this project suggest that Alzheimer’s disease is better understood as a multi layer system instability rather than a condition driven by a single pathway. By integrating neuronal signaling, gene expression, and network level dysfunction, this work provides a more complete explanation of how early biological changes lead to long term cognitive decline.

A key conclusion is that disruptions in NMDAR signaling, calcium regulation, and gene expression such as BACE1 interact and amplify each other over time. This indicates that Alzheimer’s progression is driven by interconnected failures rather than a single cause. From this, I learned that targeting only one component, such as amyloid buildup, may not be sufficient to stop disease progression.

This challenges current approaches that focus on masking symptoms or targeting late stage effects, which often provide only temporary relief. In contrast, this project supports identifying and stabilizing early system level changes before irreversible damage occurs.

A mechanism based comparison with existing treatments, including Donepezil, Memantine, and amyloid targeting therapies, shows that these approaches act on isolated pathways or late stage processes. In contrast, PER operates across multiple biological levels, including neuronal signaling, gene expression, and cellular stability, allowing it to address upstream drivers rather than downstream effects. Overall, this work introduces a direction focused on root-cause intervention and long term system stability rather than short term symptom management, while recognizing the need for further validation.

To extend this project, future work will focus on refining PER by expanding gene targets and optimizing pathway regulation. Improved simulation models could better predict long term system responses across Alzheimer’s subtypes. Future validation would include in vitro neuronal cell studies to assess how gene modulation affects pathways such as BACE1 expression and calcium signaling. DeepNeuroGenX could be improved by increasing data diversity and adding real time monitoring for continuous risk assessment. A limitation is reliance on modeled relationships, and future work should improve accuracy and account for variability between individuals.

There are many people I would like to thank, especially my family. My sister sparked my interest in science fairs at a young age, and getting this far feels like a childhood dream. When I was stressed from presenting, she made me laugh and gave helpful advice. I would also like to thank my mom, who picked me up from school to practice and introduced me to biology early on. She always kept me motivated, my dad for getting materials and supporting me, even while facing his own health challenges. I am also grateful to my friends, especially Aashi and Manya, who supported me after setbacks and reminded me I did my best. I would also like to thank my dog for staying up with me until 4 a.m. while I was researching. Lastly, I would like to thank everyone at ERSF, especially Ms. Joblinski and my judges.

Journal Articles

Berridge, M. J. (2010). Calcium hypothesis of Alzheimer’s disease. Pflügers Archiv - European Journal of Physiology, 459(3), 441–449. https://doi.org/10.1007/s00424-009-0736-1

Corder, E. H., Saunders, A. M., Strittmatter, W. J., Schmechel, D. E., Gaskell, P. C., Small, G. W., Roses, A. D., Haines, J. L., & Pericak-Vance, M. A. (1994). Gene dose of apolipoprotein E type 4 allele and the risk of Alzheimer’s disease in late onset families. Science, 261(5123), 921–923. https://doi.org/10.1126/science.8346443

De Strooper, B., & Karran, E. (2016). The cellular phase of Alzheimer’s disease. Cell, 164(4), 603–615. https://doi.org/10.1016/j.cell.2015.12.056

Frisoni, G. B., Fox, N. C., Jack, C. R., Scheltens, P., & Thompson, P. M. (2010). The clinical use of structural MRI in Alzheimer disease. Nature Reviews Neurology, 6(2), 67–77. https://doi.org/10.1038/nrneurol.2009.215

Hardy, J., & Higgins, G. (1992). Alzheimer’s disease: The amyloid cascade hypothesis. Science, 256(5054), 184–185. https://doi.org/10.1126/science.1566067

Huang, Y., & Mahley, R. W. (2014). Apolipoprotein E: Structure and function in lipid metabolism, neurobiology, and Alzheimer’s disease. Neuron, 76(4), 871–885. https://doi.org/10.1016/j.neuron.2012.11.026

Jack, C. R., Bennett, D. A., Blennow, K., Carrillo, M. C., Dunn, B., Haeberlein, S. B., Holtzman, D. M., Jagust, W., Jessen, F., Karlawish, J., Liu, E., Molinuevo, J. L., Montine, T., Phelps, C., Rankin, K., Rowe, C. C., Scheltens, P., Siemers, E., Snyder, H. M., & Sperling, R. (2018). NIA-AA research framework: Toward a biological definition of Alzheimer’s disease. Alzheimer’s & Dementia, 14(4), 535–562. https://doi.org/10.1016/j.jalz.2018.02.018

Karran, E., & De Strooper, B. (2022). The amyloid cascade hypothesis: Are we poised for success? Nature Reviews Drug Discovery, 21(5), 306–318. https://doi.org/10.1038/s41573-021-00335-6

Liu, C. C., Kanekiyo, T., Xu, H., & Bu, G. (2013). Apolipoprotein E and Alzheimer disease: Risk, mechanisms and therapy. Nature Reviews Neurology, 9(2), 106–118. https://doi.org/10.1038/nrneurol.2012.263

Mahley, R. W., & Huang, Y. (2012). Small-molecule structure correctors target abnormal protein structure and function. Journal of Medicinal Chemistry, 55(21), 8997–9008. https://doi.org/10.1021/jm3008618

Palop, J. J., & Mucke, L. (2016). Network abnormalities and interneuron dysfunction in Alzheimer disease. Nature Reviews Neuroscience, 17(12), 777–792. https://doi.org/10.1038/nrn.2016.141

Selkoe, D. J., & Hardy, J. (2016). The amyloid hypothesis of Alzheimer’s disease at 25 years. EMBO Molecular Medicine, 8(6), 595–608. https://doi.org/10.15252/emmm.201606210

Subramanian, J., & Tremblay, M.-È. (2021). Synaptic loss and neurodegeneration. Frontiers in Cellular Neuroscience, 15, 681029. https://doi.org/10.3389/fncel.2021.681029

Vassar, R., Kuhn, P.-H., Haass, C., Kennedy, M. E., Rajendran, L., Wong, P. C., & Lichtenthaler, S. F. (2009). Function, therapeutic potential and cell biology of BACE proteases: Current status and future prospects. Journal of Neurochemistry, 109(1), 47–61. https://doi.org/10.1111/j.1471-4159.2009.05946.x

Zhang, B., Gaiteri, C., Bodea, L.-G., Wang, Z., McElwee, J., Podtelezhnikov, A. A., Zhang, C., Xie, T., Tran, L., Dobrin, R., Fluder, E., Clurman, B., Melquist, S., Narayanan, M., Suver, C., Shah, H., Mahajan, M., Gillis, T., Mysore, J., … Emilsson, V. (2013). Integrated systems approach to Alzheimer’s disease. Cell, 153(3), 707–720. https://doi.org/10.1016/j.cell.2013.03.030

Books

Alberts, B., Johnson, A., Lewis, J., Morgan, D., Raff, M., Roberts, K., & Walter, P. (2022). Molecular biology of the cell(7th ed.). Garland Science.

Bear, M. F., Connors, B. W., & Paradiso, M. A. (2020). Neuroscience: Exploring the brain (4th ed.). Wolters Kluwer.

Goedert, M., & Spillantini, M. G. (2019). Neurodegeneration and Alzheimer’s disease. Cambridge University Press.

Kandel, E. R., Schwartz, J. H., Jessell, T. M., Siegelbaum, S. A., & Hudspeth, A. J. (2013). Principles of neural science(5th ed.). McGraw-Hill Education.

Purves, D., Augustine, G. J., Fitzpatrick, D., Hall, W. C., LaMantia, A.-S., Mooney, R., Platt, M. L., & White, L. E. (2018). Neuroscience (6th ed.). Oxford University Press.

Images / Figures

Eye Solutions. (n.d.). Fundus photography eye test [Image]. https://www.eyesolutions.in/retina/fundus-photography-eye-test/

ResearchGate. (2015). BACE1 is regulated at various stages of expression [Figure]. https://www.researchgate.net/figure/Fig-2-BACE1-is-regulated-at-various-stages-of-expression-Postive-and-negative_fig2_284517307

ResearchGate. (2020). Single nucleotide polymorphisms (SNPs) [Figure]. https://www.researchgate.net/figure/Single-nucleotide-polymorphisms-SNPs-are-genetic-mutations-that-alter-single-base-in_fig2_338958995

ResearchGate. (2023a). Early signs of mild cognitive impairment [Figure]. https://www.researchgate.net/figure/Early-signs-of-mild-cognitive-impairment_fig1_371811860

ResearchGate. (2023b). Epigenetic modulation system [Figure]. https://www.researchgate.net/figure/nfluences-of-the-epigenetic-modulation-system-From-Ref-5_fig1_374824079

ResearchGate. (2024). Perceived 10-year risk of dementia [Figure]. https://www.researchgate.net/figure/Perceived-10-year-risk-of-dementia-compared-to-personal-thresholds-at-which-one-considers_fig2_391677998

Webpages

AZo Life Sciences. (2023). What is antisense technology? https://www.azolifesciences.com/article/What-is-Antisense-Technology.aspx

Estuary. (2023). Data normalization explained. https://estuary.dev/blog/data-normalization/

FreeCodeCamp. (2023). How to calculate definite and indefinite integrals in Python. https://www.freecodecamp.org/news/calculate-definite-indefinite-integrals-in-python/

GeeksforGeeks. (2023a). Feature selection vs feature extraction. https://www.geeksforgeeks.org/machine-learning/feature-selection-vs-feature-extraction/

GeeksforGeeks. (2023b). Implementation of XGBoost (extreme gradient boosting). https://www.geeksforgeeks.org/machine-learning/implementation-of-xgboost-extreme-gradient-boosting/

IBM. (2022). What is machine learning? https://www.ibm.com/topics/machine-learning

Mayo Clinic. (2022). Alzheimer’s disease biomarkers. https://www.mayoclinic.org

National Institutes of Health. (2023). Artificial intelligence in medicine. https://www.nih.gov

Statista. (2024). Number of deaths from Alzheimer’s disease per 100,000 people in the U.S. https://www.statista.com/chart/30884/number-of-deaths-from-alzheimers-disease-per-100000-people-in-the-us/

V7 Labs. (2023). Data preprocessing guide. https://www.v7labs.com/blog/data-preprocessing-guide

Vancouver Coastal Health Research Institute. (2021). Road map for early brain disease detection. https://www.vchri.ca/stories/2021/07/10/road-map-early-brain-disease-detection

Whatagraph. (2023). Custom reporting software. https://whatagraph.com/custom-reporting-software

World Health Organization. (2023). Dementia. https://www.who.int/news-room/fact-sheets/detail/dementia