An interactive, web-based tool for iteratively determining optimal experimental parameters using Bayesian optimization. Experiments can be downloaded or resumed at any time using their unique identifier
Optimize your experiment here
We extracted all materials science related concepts from academic literature to create a "Map of Materials Science". Stay tuned for more..!
See the 'Materials Science Map'
We are collecting experimental powder X-ray diffraction data to create a free high-quality dataset with a broad range of samples and materials. Please visit our website in order to contribute your data and to learn more about becoming a co-author in the associated paper.
Contribute here
MeganExplains features self-explaining graph neural network models that can predict various molecular properties in a transparent fashion. Besides the pure predictions, xAI practices such as saliency maps and counterfactuals are used to provide insight to the model's internal reasoning process
Try it out yourself
How toxic is your favourite food additive? Don't know? How about we find out! We're taking you on a journey through molecular science and AI, using innovative, new technology to find answers to the real questions, like "Is water a metal?" or "How flammable is my neighbour?"
Visit our website to begin
A tool to predict the synthesis parameters of new MOFs with promising predicted properties. Our ML model is trained on a new MOF synthesis database based on automatic extraction of synthesis parameters from the literature. Even at an initial stage, it exhibits a good prediction performance, outperforming human expert predictions.
Predict MOF synthesis conditions here