Topp Beta-Cell Mass / Insulin / Glucose Model

Where this model is curated

The entry card says what the model is for: which mechanism nodes it links to, how faithfully, and what it found. This page holds the model itself.

disorder Type 2 Diabetes Mellitus kb/disorders/Type_2_Diabetes_Mellitus.yaml

Committed run

From exports/model_runs/BIOMD0000000341.json, regenerated by just gen-model-results. Scenario roots link to the mechanism node they drive on the entry page.

Simulation results — 15 scenarios run with dismech-perturb (tellurium / libRoadRunner CVODE) over 4000 h; baseline si = 0.72

Interpretation caveat. This model is bistable and its baseline sits near the saddle separating the physiological and pathological fixed points, so every impairing lesion collapses to the same attractor (glucose 600 mg/dL, beta-cell mass 0) regardless of magnitude. Scenario outcomes are therefore NOT comparable in severity, and severity tiers are suppressed for this model. Two scenarios in particular must not be read as clinical claims: GCK loss-of-function is GCK-MODY (MODY2), which causes mild, non-progressive fasting hyperglycemia with no beta-cell mass loss and usually needs no pharmacotherapy; and the TCF7L2 risk allele is a modest susceptibility factor (per-allele odds ratio ~1.4), not a deterministic route to beta-cell failure.

Scenario	Beta_Cell_Mass (mg)	Fasting_Plasma_Glucose (mg/dL)	Plasma_Insulin (uU/mL)	Activated phenotypes
Healthy baseline si = 0.72	300.0	100.0	10.0	reference
Severe insulin resistance (si=0.30) drives Insulin Resistance	0.0 0.00× baseline	600.0 6.00× baseline	0.0 0.00× baseline	Hyperglycemia
Insulin resistance (si=0.45) drives Insulin Resistance	0.0 0.00× baseline	600.0 6.00× baseline	0.0 0.00× baseline	Hyperglycemia
PPARG loss-of-function (insulin resistance) drives Insulin Resistance	0.0 0.00× baseline	600.0 6.00× baseline	0.0 0.00× baseline	Hyperglycemia
TCF7L2 risk variant (impaired secretion) drives Beta Cell Dysfunction	0.0 0.00× baseline	600.0 6.00× baseline	0.0 0.00× baseline	Hyperglycemia
KCNJ11 K_ATP gain (impaired secretion) drives Beta Cell Dysfunction	0.0 0.00× baseline	600.0 6.00× baseline	0.0 0.00× baseline	Hyperglycemia
HNF1A loss-of-function (MODY3) drives Beta Cell Dysfunction	0.0 0.00× baseline	600.0 6.00× baseline	0.0 0.00× baseline	Hyperglycemia
GCK loss-of-function (glucose-sensing defect) drives Beta Cell Dysfunction	0.0 0.00× baseline	600.0 6.00× baseline	0.0 0.00× baseline	Hyperglycemia
SLC5A2 loss-of-function (renal glucosuria; protective) drives Hyperglycemia	210.0 0.70× baseline	100.0 1.00× baseline	7.0 0.70× baseline	none
Metformin (reduced hepatic glucose output) drives Hepatic Glucose Overproduction	307.2 1.02× baseline	100.0 1.00× baseline	10.24 1.02× baseline	none
Thiazolidinedione (insulin sensitizer) drives Insulin Resistance	300.0 1.00× baseline	100.0 1.00× baseline	10.0 1.00× baseline	none
SGLT2 inhibitor (insulin-independent glucose clearance) drives Hyperglycemia	504.0 1.68× baseline	100.0 1.00× baseline	16.8 1.68× baseline	none
Sulfonylurea (secretagogue) - fails once beta cells collapse drives Beta Cell Dysfunction	0.0 0.00× baseline	600.0 6.00× baseline	0.0 0.00× baseline	Hyperglycemia
GLP-1 receptor agonist (secretion + reduced hepatic output) drives Incretin Axis Dysfunction	260.571429 0.87× baseline	100.0 1.00× baseline	12.16 1.22× baseline	none
Insulin therapy (increased net insulin action) drives Hyperglycemia	240.0 0.80× baseline	100.0 1.00× baseline	8.0 0.80× baseline	none
Metformin + SGLT2 inhibitor (severe disease) drives Hyperglycemia	244.8 0.82× baseline	100.0 1.00× baseline	8.16 0.82× baseline	none

Derived artifact, regenerated by just gen-model-results — not curated evidence. Values rounded to 6 decimals. Config e8ecf0513c19, model.xml e86507d1dc05. Phenotypes activate per the curated variable thresholds.

Phenotype thresholds

Phenotype	Variable	Direction	Threshold	Reading
Hyperglycemia	`Fasting_Plasma_Glucose`	above	126.0	absolute
Hyperinsulinemia	`Plasma_Insulin`	above	18.0	absolute

Gene effects

How config.yaml turns a gene perturbation into a model parameter change: the multiplier applied for a loss- or gain-of-function.

Gene	Parameter	Effect	Rationale
`PPARG`	`si`	LoF ×0.4	PPARG is the master adipogenic regulator and the thiazolidinedione target; loss-of-function (e.g. FPLD3) causes severe insulin resistance. Modeled as reduced whole-body insulin sensitivity si.
`TCF7L2`	`sigma`	LoF ×0.5	TCF7L2 is the strongest common type 2 diabetes risk locus; risk alleles impair glucose-stimulated insulin secretion. Modeled as reduced maximal per-beta-cell secretory capacity sigma.
`KCNJ11`	`sigma`	LoF ×0.5	KCNJ11 encodes the Kir6.2 subunit of the beta-cell K_ATP channel; activating variants hold the channel open and blunt insulin secretion (neonatal diabetes / T2D risk). Modeled as reduced secretory capacity sigma.
`HNF1A`	`sigma`	LoF ×0.5	HNF1A loss-of-function causes MODY3, a progressive beta-cell secretory defect. Modeled as reduced secretory capacity sigma.
`GCK`	`alpha`	LoF ×3.0	Glucokinase is the beta-cell glucose sensor; loss-of-function raises the glucose set-point for insulin secretion (MODY2 when heterozygous; permanent neonatal diabetes when biallelic). Modeled as an increased half-maximal glucose constant alpha (higher glucose needed for the same secretion).
`SLC5A2`	`Eg0`	LoF ×2.5	SLC5A2 encodes SGLT2, the renal glucose transporter and gliflozin target; loss-of-function causes familial renal glucosuria and raises insulin- independent glucose clearance. Modeled as increased glucose effectiveness Eg0 (protective).

Run it locally

uv run python -m dismech.perturb kb/disorders/Type_2_Diabetes_Mellitus.yaml --alljust gen-model-results --id BIOMD0000000341

Files

config.yaml model.xml