根因医学

On the Computational Framework for Root Cause Medicine

Based on capomics theory, proposing a three-layer computational framework for root cause medicine: phenotype layer, functional layer, and root cause layer.

熊江辉 · 2025-10-16
2

On the Computational Framework for Root Cause Medicine

Computational Framework for Root Cause Medicine

Author: Xiong Jianghui

3

The Problem: The Medical Paradigm Dilemma from "Treating Symptoms" to "Treating Root Causes"

Modern medicine faces a profound paradox: despite enormous advances in diagnostic technology, drug development, and clinical intervention, treatment outcomes for complex diseases remain disappointing. As noted in a paper (Network pharmacology: curing causal mechanisms instead of treating symptoms, Trends Pharmacol Sci. 2022 Feb;43(2):136-150), most drugs have extremely low efficacy against complex diseases, and drug discovery success rates continue to decline. The root of this dilemma lies in the current medical paradigm of "organ-centrism" and the dogmatic thinking of "one disease, one target, one drug."

The traditional medical model treats disease as isolated organ pathology, with treatment strategies focused on symptom relief rather than correction of causal mechanisms. This approach ignores the systemic nature of disease — most chronic diseases are not failures of a single organ, but comprehensive manifestations of multi-level, multi-system dysfunction. The rise of network pharmacology and systems medicine is precisely a response to this dilemma: we need to replace descriptive disease phenotypes with causal, multi-target signaling modules to achieve precise and effective therapeutic interventions.

In the fields of functional medicine and longevity medicine, many scholars and clinical experts have raised the issue of identifying the root causes of diseases and phenotypes, considering this a common and consensual need for the next generation of medical systems. However, the theoretical foundations for root cause analysis — especially practical tools and computational frameworks — remain lacking in systematicity.

Facing multi-omics data, functional medicine and longevity medicine practice has gradually accumulated vast amounts of genomic, proteomic, metabolomic, exposomic, biological age, and epigenetic DNA methylation data. How to integrate these data, discover root causes, and identify intervention targets is an enormous challenge. Many experts place their hopes on AI development, expecting AI to aggregate and integrate massive data and generate insights.

I believe the above approach follows inductive reasoning — the approach of experimental science. To solve the problem of multi-omics data integration and root cause discovery, we may need to shift perspective — abandon induction, turn to deduction, and starting from first principles, establish a general theoretical and computational framework. Therefore, based on my proposed Capomics theory, I now propose a three-layer computational framework for Root Cause Medicine — the phenotype layer, the functional layer, and the root cause layer. This article will systematically elucidate the rationality of this architecture, its practical value, and why specific concepts should be assigned to their respective layers.

4

First Principles: From "Structure-Function-Capacity Reserve" to the Three-Layer Architecture

The core idea of Capomics theory can be summarized as: structure determines function, function generates capability, and capacity reserve determines health status. This idea permeates every scale of the living system — from molecules, cells, and tissues to organs, systems, and the whole organism.

- Structural level: DNA sequences, protein folding, organelle morphology, organ anatomical structure

- Functional level: Gene expression, enzymatic reactions, mitochondrial energy production, cardiac pumping

- Capacity Reserve: DNA repair capacity, antioxidant capacity, metabolic adaptation capacity, organ functional reserve

The essence of disease is the depletion of capacity reserve. Young and healthy individuals possess ample capacity reserves to cope with various internal and external stresses; disease states occur when capacity reserves fall below critical thresholds and the system can no longer maintain homeostatic balance.

The three-layer architecture of Root Cause Medicine is constructed based on this first principle:

- Phenotype layer corresponds to "the external manifestations after capacity reserve depletion"

- Functional layer corresponds to "the intermediate process of capacity reserve decline"

- Root cause layer corresponds to "the underlying mechanisms leading to capacity reserve depletion"

5

Theoretical Rationality of the Three-Layer Architecture

Why are the phenotype layer, functional layer, and root cause layer all indispensable? These three layers are not arbitrarily divided, but represent the natural hierarchy of the disease causal chain:

1. The Necessity of the Phenotype Layer: Observability and the Starting Point of Clinical Practice

The phenotype layer is the "entry point" of medical practice. When patients present to the clinic, they manifest phenotypic abnormalities — chest pain, elevated blood glucose, masses, etc. The characteristics of this layer are:

- Directly observable: Obtainable without complex technology (physical examination, routine lab tests)

- Clinical decision basis: Diagnosis, staging, and treatment evaluation are all based on phenotypic indicators

- Strong patient perception: Phenotypes directly affect quality of life

However, the limitation of the phenotype layer lies in its non-specificity and滞后性 (latency). The same phenotype (e.g., fatigue) may arise from dozens of different functional disorders; by the time a phenotype manifests, underlying damage has often been ongoing for years. Therefore, intervening only at the phenotype layer is "treating the head when the head aches, treating the foot when the foot hurts" — it cannot halt disease progression.

2. The Criticality of the Functional Layer: The Bridge Connecting Phenotypes and Root Causes

The functional layer reveals "why these phenotypes appear." It represents the capacity decline of organs and systems:

- Organ aging: Declining cardiac ejection fraction, decreasing glomerular filtration rate, reduced lung capacity

- Metabolic aging: Declining mitochondrial ATP production, glucose-lipid metabolism disorders, redox imbalance

- Immune dysregulation: Chronic low-grade inflammation, autoimmune reactions, immunosenescence

- Neurodegeneration: Declining synaptic plasticity, neurotransmitter imbalance, insufficient cerebral blood perfusion

Why place these concepts in the functional layer? Because they are the combined result of multiple root causes and the common source of multiple phenotypes.

The functional layer possesses the network characteristics of "multiple causes, one effect" and "one cause, multiple effects", and is a critical node in the disease causal network. Intervening at the functional layer often produces systemic improvements rather than relief of a single symptom.

3. The Determinativeness of the Root Cause Layer: The Ultimate Drivers of Disease

The root cause layer answers "why functions decline." This layer contains the most fundamental molecular-cellular mechanisms, which are the "prime movers" of disease occurrence.

Why choose Aging Hallmarks and Cancer Hallmarks as the core content of the root cause layer? This is based on the following considerations:

The Universality of Aging Hallmarks:

The nine hallmarks of aging summarized by López-Otín et al. (Cell, 2013, 2023) (now expanded to twelve hallmarks) represent the common underlying mechanisms of virtually all age-related diseases:

- Genomic instability (accumulation of DNA damage)

- Telomere attrition (limitation on cell division count)

- Epigenetic alterations (dysregulation of gene expression control)

- Loss of proteostasis (accumulation of misfolded proteins)

- Deregulated nutrient sensing (abnormalities in mTOR, AMPK, Sirtuins pathways)

- Mitochondrial dysfunction (decline in energy metabolism)

- Cellular senescence (accumulation of senescent cells and SASP secretion)

- Stem cell exhaustion (decline in regenerative capacity)

- Altered intercellular communication (dysregulation of neuro-immune-endocrine networks)

These mechanisms are not unique to any single disease, but are common roots of almost all chronic diseases including cardiovascular disease, diabetes, neurodegenerative diseases, and cancer. They are interrelated and mutually causal, forming a complex root cause network.

The Specificity of Cancer Hallmarks:

The ten hallmarks of cancer proposed by Hanahan and Weinberg represent the key steps in the transformation of cells from normal to malignant. Cancer is an "extreme form" of aging — when cells evade senescence, achieve immortality, and reprogram metabolism, the transition from simple functional decline to malignant proliferation occurs.

Placing aging hallmarks and cancer hallmarks in the root cause layer is justified because they satisfy three criteria for "root causes":

- Causality: They are direct causes of functional disorders, not consequences

- Universality: They apply to multiple diseases, not just specific diseases

- Intervenability: Intervention strategies targeting these mechanisms exist or are under development (e.g., senolytics, telomerase activators, NAD+ supplementation, etc.)

6

Practical Value of the Three-Layer Architecture

1. Guiding Precision Diagnosis: From "Symptom Matching" to "Mechanism Tracing"

Traditional diagnosis relies on "symptom-disease" matching: high blood glucose → diabetes, chest pain → coronary heart disease. The problems with this diagnostic approach are:

- Ignoring individual differences (patients with the same diagnosis may have entirely different root causes)

- Inability to predict complications (not knowing which patients will progress to renal failure, blindness, etc.)

- Uniform treatment protocols (same treatment for the same diagnosis)

The three-layer architecture supports a "reverse causal reasoning" diagnostic model:

Phenotype data → Functional assessment → Root cause tracing → Individualized diagnosis

2. Guiding Stratified Intervention: Multi-level, Multi-target Treatment Strategies

The three-layer architecture supports a "stratified synergistic intervention" model:

Phenotype layer intervention: Rapid symptom relief, improved quality of life

- Glucose-lowering drugs to control blood sugar, antihypertensive drugs to control blood pressure

- Applicable in acute phases, severe symptom phases

- Limitation: does not alter disease course, relapse likely upon discontinuation

Functional layer intervention: Restore systemic capabilities, delay disease progression

- Enhance immune function (vitamin D, zinc, probiotics)

- Applicable in chronic phases, disease progression phases

- Advantage: simultaneous improvement of multiple phenotypes, reduced complications

Root cause layer intervention: Reverse underlying mechanisms, achieve "cure"

- Clear senescent cells (Senolytics such as Dasatinib + Quercetin)

- Activate autophagy (intermittent fasting, rapamycin analogues)

- Supplement NAD+ (NMN, NR to reverse mitochondrial and DNA repair function)

- Epigenetic reprogramming (Yamanaka factors, DNA methylation modulation)

- Applicable for early prevention, fundamental cure

- Challenge: technically complex, long-term safety needs verification

Best Practice: Three-Layer Synergy

- Acute phase: Phenotype layer intervention as primary (rapid symptom control)

- Stable phase: Functional layer intervention as primary + phenotype layer maintenance

- Recovery phase: Root cause layer intervention as primary + functional layer support

3. Guiding Preventive Medicine: From "Early Detection" to "Early Intervention"

Traditional preventive medicine emphasizes "early screening, early diagnosis," but often by the time phenotypes appear (e.g., elevated tumor markers, abnormal blood glucose), damage is already irreversible.

The three-layer architecture supports "advancing the intervention window":

- Root cause layer monitoring: When function and phenotype are normal, detect aging biomarkers (e.g., DNA methylation age, telomere length, senescent cell markers p16/p21)

- Functional layer assessment: When phenotypes are normal, assess organ functional reserves (e.g., cardiopulmonary exercise testing, cognitive reserve testing)

- Ultra-early intervention: When root cause or functional layer abnormalities emerge but phenotypes remain normal, begin intervention immediately

This "advancement" strategy could move the intervention window forward by 10-20 years, fundamentally altering disease natural history.

4. Guiding Aging Research: A Unified Framework for Integrating Fragmented Knowledge

The current field of aging research suffers from severe fragmentation: telomere researchers, mitochondrial researchers, inflammation researchers, and epigeneticists each work independently, lacking a common language and integrative framework.

The three-layer architecture provides a unified causal hierarchy:

- All aging mechanisms (root cause layer) → All functional decline (functional layer) → All disease phenotypes (phenotype layer)

- Interactions between different root causes can be quantitatively modeled

- The impact of intervening on one root cause on other layers can be predictively assessed

This enables us to construct a panoramic map of "aging-disease", guiding the development of systematic anti-aging strategies.

7

Why Are These Concepts Placed in Their Respective Layers? Criteria for Layer Assignment

How to judge which layer a biological phenomenon or clinical indicator should belong to? I propose the following criteria:

Criterion 1: Causal Position

- Root cause layer: Is the "cause" of other phenomena, rarely the "effect" itself

- Functional layer: Is both "effect" (caused by root causes) and "cause" (leading to phenotypes)

- Phenotype layer: Primarily "effect," rarely "cause"

Criterion 2: Universality/Specificity

- Root cause layer: Highly universal, applicable to multiple diseases

- Functional layer: Moderately universal, applicable to multiple diseases of a specific system or organ

- Phenotype layer: Highly specific, often disease-specific

Example: Why does "cellular senescence" belong to the root cause layer?

Cellular senescence and its secreted SASP factors are involved in virtually all age-related diseases — atherosclerosis, osteoarthritis, neurodegeneration, cancer, etc. Its high universality makes it a root cause.

Criterion 3: Measurability and Clinical Accessibility

- Phenotype layer: Easily measurable, routinely accessible in clinical practice (e.g., blood pressure, blood glucose)

- Functional layer: Requires specialized assessment (e.g., cardiopulmonary exercise testing, cognitive testing)

- Root cause layer: Requires advanced technology (e.g., genomic sequencing, epigenetic analysis)

This criterion also explains why medical practice often remains at the phenotype layer — because it is the easiest to measure. But as technology advances (e.g., portable genetic testing, rapid aging biomarker detection), root cause layer testing is becoming more accessible, which will drive the transition from phenotype medicine to root cause medicine.

Criterion 4: Time Scale of Intervention

- Phenotype layer intervention: Rapid effect (hours to days), but easy relapse

- Functional layer intervention: Moderate effect onset (weeks to months), sustained effects

- Root cause layer intervention: Slow effect onset (months to years), but may reverse disease

Example: Three-Layer Intervention for Blood Glucose Control

- Phenotype layer: Insulin injection, lowering blood glucose within 2 hours (but not changing insulin resistance)

- Functional layer: Supplement CoQ10 to improve mitochondrial function, insulin sensitivity improves after 4-8 weeks

- Root cause layer: Intermittent fasting induces autophagy, clears damaged mitochondria, pancreatic islet function partially recovers after 3-6 months

8

Conclusion: From Medical Philosophy to Clinical Practice

The three-layer architecture of Root Cause Medicine is not merely an analytical framework, but a shift in medical philosophy:

- From reductionism to systems theory: Disease is not a failure of an isolated organ, but an imbalance of multi-level networks

- From static to dynamic: Health and disease are dynamic balances of capacity reserve, not binary opposites

- From treatment to prevention: Intervening at the root cause layer may prevent disease decades before phenotypes appear

- From extending life to improving quality: The goal is not simply to live longer, but to maintain high capacity reserves, achieving "healthy aging"

The three-layer architecture proposed by Capomics theory — phenotype layer, functional layer, and root cause layer — provides an actionable theoretical foundation and computational framework for this transformation. By systematically tracing the causal chain from root causes to phenotypes, we can:

- More accurately diagnose the essential causes of disease

- More precisely design individualized intervention plans

- More effectively predict and prevent disease progression

- More scientifically evaluate intervention effects and optimize strategies

This is not a distant vision of future medicine, but a realistic path that can be put into practice under current technological conditions. With the popularization of multi-omics technology, the development of artificial intelligence, and breakthroughs in geroscience, Root Cause Medicine will inevitably move from theory to clinical practice, from a few cutting-edge centers to universal healthcare.

I believe that in the next 10-20 years, we will witness a profound transformation of medicine: hospitals will no longer merely be places for treating diseases, but centers for assessing capacity reserves and optimizing life systems; doctors will no longer merely be prescribers, but consultants for capacity management; patients will no longer be passive recipients, but active decision-makers managing their own health.

This is the ultimate vision of Root Cause Medicine — enabling everyone to understand, monitor, and optimize their own capacity reserves, thereby achieving truly autonomous health and freedom of life.