MatScreen

Find the best materials for your application. AI-powered screening with honest confidence estimates.

View on GitHub Watch Demo
230k
Materials Screened
3
Properties Predicted
5
Ensemble Models
27
Unit Tests

Features

Everything a materials scientist needs to make confident screening decisions.

Application Presets

Select "Solar Cell", "LED", "Thermoelectric", or "Wide-Gap Semiconductor" and the system configures target ranges automatically.

🎯

Calibrated Confidence

Every prediction gets an honest confidence label. Isotonic regression ensures the uncertainty bars actually mean what they claim.

Multi-Objective Ranking

Pareto sorting balances band gap, stability, and confidence simultaneously. Single-property filtering misses these tradeoffs.

📊

Property Radar Charts

Five-dimensional profiles let you compare candidates at a glance across band gap match, stability, confidence, strength, and formability.

Honest Limitations

PBE band gaps underestimate by ~40%. We label the DFT functional for each prediction. No magic claims, just transparent science.

🧪

Benchmark Validated

Forward models evaluated on Matbench v0.1 with fixed splits. Screening validated by checking known good materials appear in results.

How It Works

From target properties to ranked candidates in seconds.

1

Define your target

Specify the properties you need. "I want a material with band gap 1.0 to 1.5 eV for solar cells, thermodynamically stable."

2

Screen the database

MatScreen filters 230,000 materials from Materials Project and JARVIS, applying stability and element constraints.

3

Predict with uncertainty

An ensemble of 5 ALIGNN models predicts properties for each candidate. Disagreement between models gives calibrated confidence intervals.

4

Rank by suitability

Pareto sorting ranks candidates across all objectives simultaneously. No single property dominates. The best tradeoffs rise to the top.

5

Review with confidence

Each candidate shows predicted properties with confidence ratings. GREEN means trust it. RED means verify with DFT before acting.

HIGH CONFIDENCE

Uncertainty < 0.08 eV. The model has seen many similar materials. Prediction likely within 0.1 eV of truth.

MODERATE

Uncertainty 0.08 to 0.15 eV. Useful prediction but should be verified with a DFT calculation.

LOW CONFIDENCE

Uncertainty > 0.15 eV. Unusual material. Do not rely on this prediction alone.

Open Source

MatScreen is free and open source under the Apache 2.0 licence. Built on Materials Project, JARVIS, and ALIGNN.

View on GitHub