AmyloGraph 2.0

Expanding the evidence on amyloid self-assembly modulators, and using that evidence to build prediction and generative models.

NCN · SONATA 19

From aggregation data to predictive models.

Expanding the evidence on amyloid self-assembly modulators, and using that evidence to build prediction and generative models.

Amyloids · Data curation · Machine learning

Programme: NCN · Sonata 19
Grant ID: 2023/51/D/NZ7/02847
Start date: 10 Jul 2024
Duration: 36 months
Budget: 1,797,914.00 PLN (~ €423,300.00)

The research question

AmyloGraph began as a resource for amyloid–amyloid interactions. This project extends its scope to a wider range of self-assembly modulators and the experimental context in which they act. The resulting curated evidence supports models that go beyond predicting whether a protein aggregates, towards understanding how its aggregation can be changed.

What the project sets out to do

01 / RESEARCH AIM

Expand the evidence base

Extend AmyloGraph to non-protein modulators, including experimental conditions and other essential metadata.

02 / RESEARCH AIM

Predict modulation

Train machine learning models to predict how a modulator affects amyloid aggregation.

03 / RESEARCH AIM

Explore new modulators

Develop generative models and assess proposed modulators computationally and experimentally.

People & support

Project lead: Michał Burdukiewicz.

National Science Center Poland

Full project title
Taming aggregation with AmyloGraph 2.0: a database and prediction models of amyloid self‑assembly modulators

Explore further

Explore the AmyloGraph database ↗