NUDGE (nudge-bio)

Node/edge Ultrasensitivity Diagnostic for Gene-regulatory Effects.

NUDGE is a mechanism-attribution tool for perturbation (Perturb-seq) screens. It fits a compositional, differentiable gene-regulatory circuit model to single-cell count data and classifies each perturbation by mechanism — does a knockdown move a switch's threshold (K), change its gain (n), or shift its ceiling (v_max)? — a distinction the field's default linear models cannot make. Its defining property is honesty: when the data cannot identify the mechanism, NUDGE abstains (unresolved / off-model) rather than emit a confident guess. Built on MADDENING, a differentiable JAX graph-physics engine.

Gene-regulatory circuits are what NUDGE was initially built for — and what it's named for — but that was only the first target. Its core is domain-general: a compositional, differentiable ODE model plus a calibrated abstention gate. Anywhere a mechanism hides in the shape of a dynamical response and a naive fit would over-claim, the same engine applies — and it already reaches well beyond gene circuits to microbial community dynamics (generalized Lotka–Volterra trajectory fits, NUDGE-METHOD-012), protein aggregation kinetics (amyloid fibrillization, NUDGE-METHOD-013), and differentiable experimental design (gradient-optimal experiments + matrix-free identifiability at scale, NUDGE-METHOD-014). Each new domain is a new module over a frozen core — never a rewrite.

NUDGE originated at the Built with Claude: Life Sciences hackathon (July 2026), and is itself an experiment in Claude-assisted development — the git history is written to make that involvement auditable.

Install

pip install nudge-bio

Core install pulls maddening[ift]>=0.3.1, jax==0.5.1 (pinned), numpy, optax, pydantic, anndata, typer, pyyaml. Python ≥ 3.10.

Optional extras:

Quickstart

Attribute a mechanism from a dose-response curve

The flagship positive: give NUDGE a knockdown dose-response of a readout signature and it calls switch vs graded (or abstains). Here a genuinely ultrasensitive curve resolves to switch:

import numpy as np
from nudge.mechanisms.regulatory import hill_repression
from nudge.inference.dose_response import fit_dose_response, classify_dose_response

# A knockdown dose-response of a self-renewal signature (an ultrasensitive switch, n=6).
dose = np.linspace(0.0, 1.0, 22)
response = 0.2 + np.asarray(hill_repression(dose, 0.5, 6.0, 0.8))
response += np.random.default_rng(0).normal(0.0, 0.02, dose.shape)  # measurement noise

fit = fit_dose_response(dose, response, direction="repress", n_boot=200)
call, reason = classify_dose_response(fit)
print(f"call = {call!r}")
print(f"apparent gain n = {fit.n:.1f}  (95% CI {fit.ci_n[0]:.1f}-{fit.ci_n[1]:.1f})   "
      f"K = {fit.k_threshold:.2f}   R2 = {fit.r2:.2f}")

call = 'switch'
apparent gain n = 6.5  (95% CI 6.0-7.5)   K = 0.49   R2 = 1.00

(n is an apparent population gain, not molecular cooperativity — NUDGE says so in the reason string.) A curve whose doses don't span the inflection, or whose gain CI straddles the ultrasensitive line, returns unresolved / no-effect instead.

The two verbs — fit and design

result = nudge.fit(adata, circuit)     # → MechanismMap (per-perturbation calls + uncertainty)
plan   = nudge.design(target)          # → ranked interventions, behind safety gates

fit wants raw integer counts — NUDGE owns the observation model (a negative-binomial + dropout count model; the mechanism signal lives in the shape of the single-cell distribution, which standard log/normalize pipelines destroy). Pass an AnnData of raw counts with obs["condition"] labels (a "WT" control plus one label per perturbation).

Honesty, by design: a single under-powered snapshot at one operating point genuinely tends to abstain — the gain⇄threshold degeneracy is real, and forcing a call would be guessing. That is why the resolving capabilities read a dose axis (dose-response, above), several reporters of one latent (multi_reporter), or two operating points. On a single synthetic snapshot, nudge.fit honestly abstains:

import nudge
from nudge.circuits import ras_switch_1node
from nudge.data.synthetic import PerturbationSpec

circuit = ras_switch_1node()
adata = nudge.generate_synthetic_perturbseq(
    circuit,
    perturbations=[PerturbationSpec("KD", scope="edge", index=0, param="K", factor=3.0)],
    n_cells_per_condition=1000, seed=0,
)
result = nudge.fit(adata, circuit)     # one operating point; raw counts checked at the boundary
for c in result.calls:
    print(c.perturbation, "->", c.mechanism.value, f"(confidence {c.confidence:.2f})")

KD -> no-effect  (confidence 0.00)

That abstention is the tool working, not failing — NUDGE won't over-call a single snapshot.

Command line

nudge check-data screen.h5ad                 # raw-count guardrail — fails loudly on normalized input
nudge attribute screen.h5ad --target SOS1    # mechanism call + honest abstentions/skips
nudge explain unresolved                     # why an abstention was the honest answer
nudge mechanisms                             # the registered mechanism library + cards

What it does (capability map)

Each capability is fail-safe by construction (0 confident-wrong on its synthetic battery) and ships a Mechanism Card, tests, and a decoy it must correctly resist.

The honesty differentiator

NUDGE's whole thesis is never claim more than you measured. A confident-wrong call — a specific mechanism where the truth is "can't tell" — is the only hard failure; an abstention or a one-sided bound is a feature, not a bug. The fail-safe guarantee is measured (0% misclassification across hundreds of synthetic datasets) and adversarially red-teamed across seven rounds — every found hole was independently reproduced, then closed or locked as a regression decoy, with a final full re-scan finding 0 holes.

Learn more

• Full README, capability essay, and status: github.com/NicholasEhsanRoy/NUDGE
• Live roadmap + architecture notes: design/STATE.md
• The data contract (raw counts only, and why): docs/user_guide/data_contract.md
• Guided notebooks: dose-response flagship, inverse design, the full index
• Measured results (V&V + calibration): scripts/vv/FINDINGS.md

License

MIT. See LICENSE.