Small Molecules
(SMILES)
Supported Data
Our technology, called Small Data AI unifies 3 key capabilities: (1) structural representations of biomolecules, (2) self/semi-supervised learning from partially labelled data with less than 10k labels, and (3) motif detections in structural representations. Feel free to click on each step for further information.
Using structural representations of data enables learning in small data scenarios as structure diversity alone carries information that can be harnessed to make predictive models. Additionally, structural models can be queried for “explainability” so that our models are not black box.


DeepMirror Chem
DeepMirror Chem is our flagship app that closes the learning loop between laboratory results and experimental design. With it, our customers accelerate their small molecule drug discovery pipelines and optimisation pipelines to unravel promising leads ~4x faster. The platform helps you optimise target affinity and other drug properties (logD, hERG, ADME, etc) with active learning and our internal database.
Get in touch to request access to our app.
We simulated small molecule lead optimisation workflows both with and without DeepMirror Chem and estimate that we enable users to find the most optimal drug ~4x faster.
4x acceleration of Small Molecule lead optimisation
PREDICT
Our proprietary database and autoML infrastructure learns drug target activities from customer data and predicts common drug properties
DISCOVER
Select promising drug compounds for further testing with confidence measures and multiparameter optimisation
UNDERSTAND
Use explainable AI to unravel mechanisms of action by highlighting molecular structures that lead to high target affinity


Features
Our technology: Small Data AI




Property prediction
Interpretable AI
Confidence Estimates
Continuous Learning
Cloud Services
Intuitive Interface
Generative AI
Use certified models trained on our curated databases for property prediction (ADME, permeability, etc)
Automatically predict using the best performing deep learning models
Highlight chemical substructures that are important for predicting properties or QSAR values
Our predictions come with confidence estimates so you always know when you can trust predictions
Use multi-parameter optimisation to design experiments that bring you closer to your goal
Safe and private cloud computing (ISO27001 and SOC2 certification in progress)
Easy to use, no-code software so you don't need to know how to program to use our technology
Create small molecule for your next experiment using Generative AI (coming soon)
AutoML
Case Studies
