This blog is not being maintained anymore. Please visit docs.mqs.dk

Presentation of our Kubeflow pipelines and comparison to the Cebule API (Python SDK) for building powerful prediction pipelines.

We present a versatile graph neural network (GNN) model to predict various properties such as the HOMO-LUMO gap as well as a tutorial on training and fine-tuning custom GNN models via Cebule and/or Kubeflow.

High-throughput screening of COSMO data for thermodynamic properties.

Fine-tuning a ML Model for HOMO-LUMO gap predictions; an important property to map to a pre-liminary toxicity assessment for a wide range of chemical application domains for human and environmental safety assesments.

Develop on top of the MQS tools stack

Combining the Search Field and Search API for the high-throughput screening of molecules.

Optimized search and molecules data sheet.

High-throughput screening of quantum chemical properties of molecules.

Classical and photonic quantum computing of quantum chemistry based applications.