Disease as Regulatory Divergence: Quantifying Signal Disorder and Predicting Therapeutic Response Across Five Pathologies

Summarise with VIRAT
Authors: Duggan Matthews
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Living tissues maintain function through coordinated signalling networks spanning gene regulation, metabolism, electrophysiology, and tissue architecture. We propose that disease states can be characterised as informational divergence from healthy regulatory distributions. Using transcriptional entropy and Jensen-Shannon divergence computed from gene expression data, we develop a framework in which disease progression corresponds to increasing regulatory disorder and successful therapy corresponds to restoration of signalling coherence. We ground this empirical framework in a theoretical motivation from the companion paper series: cells are predictive systems that maintain identity by reducing divergence between internal models and environmental signals, and disease corresponds to departure from the cellular ground state under energetic constraint. The framework identifies four structurally distinct therapeutic failure modes — ignorance (insufficient signalling, treatable by metabolic support), structural blindness (failed differentiation, treatable by phase-transition therapy such as ATRA in acute promyelocytic leukaemia), incompatibility (degraded tissue architecture making cellular ground states unrealisable, treatable only by microenvironment restoration), and meta-ignorance (applying destructive interventions when the problem is coherence, treatable by paradigm shift from symptom destruction to signal restoration). We illustrate the framework across five disease domains — cancer, Alzheimer's disease, amyotrophic lateral sclerosis, autoimmune disease, and cardiovascular pathology — and identify differentiation therapy in APL as the natural experimental system. The central prediction is immediately testable: successful therapy reduces transcriptional entropy and divergence from healthy distributions prior to clinical remission. Because these quantities can be computed directly from publicly available RNA-seq datasets, the hypothesis can be evaluated within weeks without new clinical trials, new patients, or ethics approval. Falsifiable predictions are provided for each domain. If validated, divergence-based metrics may provide a general quantitative measure of biological signal disorder and a new lens for understanding disease progression and recovery. The cell that has become cancerous has not been invaded by something foreign — it has lost something essential. The cell does not need to become something new. It needs to remember what it already is.

Fields of Study & Subjects

Gene Regulatory Network Analysis Bioinformatics and Genomic Networks Gene expression and cancer classification

Publication details

Journal / source
OSF Preprints (OSF Preprints)
Publication date
20 March 2026
Language
EN
Document type
Preprint