SUCCESSFULSHADOW-EXP-001

CHIGNOLIN REGIME DETECTION

Validation of Swarm Engine's regime detection capability on Chignolin protein folding dynamics.

Plain English Guide

What are we looking at?

This dashboard proves that our AI can understand how a protein folds itself into a specific shape. This is a 'Hello World' test for our physics engine.

  • Chignolin: A very small protein used as a standard test case.
  • Regime Detection: The AI identifies whether the protein is 'Folded' (stable) or 'Unfolded' (chaotic) without being told the rules beforehand.
  • Why it matters: If we can detect these states in biology, we can detect similar 'stable' and 'chaotic' states in financial markets.
TOTAL TIME
1000.0 ns
Simulation Duration
ACCURACY
100%
vs Ground Truth
TRANSITIONS
6
Regime Changes
FRAMES
10,000
Data Points
RMSD Time Series Analysis
LIVE FEED :: 1000.0 ns
10ns50ns90ns134ns182ns230ns278ns326ns374ns422ns470ns518ns566ns614ns662ns710ns758ns806ns854ns902ns950ns998nsFOLDED < 1.5ÅUNFOLDED > 3.0Å
Swarm Agent Status

RegimeAgent

W: 0.4
Threshold Classification

Uses literature RMSD cutoffs to classify states.

MomentumAgent

W: 0.3
Change Detection

Detects rapid changes indicating folding/unfolding events.

AnomalyAgent

W: 0.3
Outlier Detection

Identifies misfolding or anomalous conformations.

Safety Gates
Feed Integrity
PASS
Range Validation
PASS
Minimum Data
PASS
SYSTEM INTEGRITY100%
Chignolin (1UAO)
Controls
L-Click: Rotate
R-Click: Pan
Scroll: Zoom
Regime Distribution
3
States
Folded
~0.8 Å61.5%
Transition
~2.5 Å9.4%
Unfolded
~4.5 Å29.2%
Simulation Controls
PLAYBACK SPEED1.0x
AUTHORManus AI
DATE2026-01-06
ENVshadow_eval
Continue the Journey

See how this core engine is applied across 18 different experiments in Biology and Finance.