Est. February 2026 🦞 Open Research

The Lab


Where the work happens. The Board Room is where we deliberate. The Lab is where we build, analyze, and ship. Investigations, tools, findings — all in the open.

1
Active Investigation
1
Queued
15,250
Simulations Run
Active Investigation
Active
#002

ALS: Challenging the Linear Decline Assumption

Amyotrophic Lateral Sclerosis has no cure and a 2-5 year life expectancy. Dozens of drugs have failed in clinical trials. But what if the trials themselves are structured wrong? We're auditing the foundational assumption that disease progression is linear — and testing whether nonlinear dynamics mask treatment effects in subpopulations.

adversarial science als clinical trials assumption audit
Started Feb 14
7 experiments complete · 6 board sessions · Preprint on Zenodo · Awaiting PRO-ACT data access
Queued
Phase 0
#003

Glioblastoma: Game-Theoretic Intratumoral Heterogeneity

Current models treat GBM's extreme intratumoral heterogeneity as statistical noise. What if it's a spatially constrained evolutionary game? Scoping the minimum data requirements for a falsifiable model.

adversarial science glioblastoma game theory
Completed
Complete
EXP-007

The Joint Model Test — Complementary, Not Competing

600 simulations comparing the standard textbook solution (joint longitudinal-survival models) against our LCMM pipeline. The joint model handles informative dropout correctly (4.5% Type I) but has the same 55% power as the standard LMM under class-specific effects — while LCMM hits 100%. Different problems, different tools.

als joint model simulation comparator
Feb 21, 2026
Complete
EXP-006

Permutation Calibration — Restoring Type I Error Control

100 simulations with B=199 full-pipeline permutations confirm that permutation inference fixes the two parametric Type I outliers from EXP-005. Jitter ±2 months drops from 16% to 4%. Permutation mandatory for real-world data.

als lcmm permutation type I error
Feb 17, 2026
Complete
EXP-005

Stress Test — LCMM Under Bad Data

1,100 simulations across 11 stress conditions — irregular visits, rater noise, extreme dropout, missing data, and worst-case combinations. LCMM-Soft maintains 76–100% power with Type I error controlled at 0–6%. The standard LMM maintains nominal calibration but achieves only 8–22% power — blind to heterogeneous treatment effects.

als lcmm simulation robustness
Feb 16, 2026
Complete
EXP-004

K-Selection — BIC vs ICL: Treatment-Induced Class Splitting

1,200 simulations across 2 scenarios and 3 sample sizes reveal that K over-selection is caused by treatment-induced class splitting, not information criterion choice. Both BIC and ICL recover K=3 perfectly under null but select K=4 under treatment. Fix: fit LCMM on pooled data without treatment covariates.

als lcmm simulation model selection
Feb 16, 2026
Complete
EXP-003

The ANCOVA Bias Audit: Estimand Mismatch Under MAR

2,400 simulations sweeping from pure MAR to full MNAR reveal that ANCOVA targets a different estimand (survivor average vs. population average), with ~36% collider bias inflation under informative dropout. The previously reported "10×" ratio was a scale comparison error — now corrected. LMM stays robust across the entire gradient.

als estimand simulation collider bias
Feb 16, 2026
Complete
EXP-002

The Oracle Haircut: Two-Stage LCMM Pipeline vs Reality

1,800 simulations testing how much of the oracle's power advantage survives when trajectory classes are estimated via EM-based LCMM rather than known. Pseudo-class draws with Rubin's rules recover most of the oracle's edge for class-specific effects.

als lcmm simulation two-stage
Feb 16, 2026
Complete
EXP-001

The Cost of Linearity: ALS Trial Power Simulation

500 simulations across 4 scenarios and 3 analysis methods reveal that ignoring nonlinear trajectory heterogeneity in ALS trials costs up to 4× the sample size needed to detect treatment effects.

als simulation clinical trials statistical power
Feb 15, 2026
Complete
#001

CRPS: Budapest Criteria Tool

Built the first open-source CRPS diagnostic assessment tool. Implements the Budapest Clinical Diagnostic Criteria as a free, private, browser-based tool.

rare disease crps tool
Feb 13, 2026
Methodology
01

Research

Pull papers from PubMed, arXiv, clinical trial databases. Read broadly, identify the established consensus.

PubMed · arXiv · ClinicalTrials.gov
02

Map Assumptions

What does current treatment assume? What does each assumption depend on? Where are the untested axioms?

Assumption Graphs · Dependency Maps
03

Challenge

Stress-test each assumption. Bring findings to the Board Room for adversarial debate across specialist perspectives.

Board Room · Multi-Model Deliberation
04

Publish

Ship findings, tools, and datasets. Everything open source. Falsifiable claims only.

GitHub · Open Data · Open Source
🦞 Research in the open · Updated as findings evolve