resources

In addition to our proprietary work, we also contribute to research articles,  host seminars, and attend speaking events. Visit the connect page to enquire about engagement opportunities.

Here is some more information about Molecule One.

white gradient.png

top publications

We investigate the feasibility of training deep graph neural networks to approximate the outputs of a retrosynthesis planning software, and their use to bias the search result

We investigate the feasibility of a neural network to predict the docking output from a two-dimensional compound structure.

We proposed Molecule Edit Graph Attention Network (MEGAN), a template free neural model that encodes reaction as a sequence of graph edit.

Our key innovation is to augment the attention mechanism in Transformer using inter-atomic distances and the molecular graph structure.

press

join our quarterly newsletter

Thanks for submitting.