Graphen Artificial Intelligence
AI of Physical Atom Networks enabling Digital Biology
Biology Central Dogma is established through DNA/RNA, which guides assembling Amino Acids to create Proteins -- the building blocks of any live. Protein is a molecule, which is a graph set of atoms.
Graphen Medical has been focusing on solving biological challenges through fundamental research utilizing graph computing and quantum physics to better predict biological functions down to the atom level.
Graphen Atom platform includes 12 tools. Most Atom tools' accuracy outperformed the best known tools in the literature, except the Protein Structure Prediction tool, which is slightly worse than the state-of-the-art but with less computational requirement.
Graphen Atom makes fundamental scientific progress down to the atom level. These tools model from proteins, to protein-protein interaction and protein-environment interaction, and to biological network and system biology viewpoints. Through scaling from points, to links, and to networks, Graphen Atom is providing advantageous performance, explainabilty and predictability. Like a map that zooms to see in different scale, Graphen Atom provides a foundation of Digital Biology and helps to predict and explain how bio mechanisms function.
Graphen Atom powers Graphen Medical's two major business units: Graphen Drugomics, which provides Drug Design services, and Graphen Genomics, which provides Consumer Genome Analysis and Precision Health solutions.
Graphen's contribution to the worldwide community on the fight with the COVID-19 coronavirus by providing our expertise in genome analysis has been widely recognized. Graphen analyzed the variants of each reported whole genome sequencing from more than 100 countries or regions to date and identified the mutations as they spread.
By understanding the mutation of each virus and locating where those variants are in its near 30,000 genetic locations, Graphen can identify virus' evolution chain and clusters. Such patterns can help better identify targets for drug development, public health prediction, or even predict the harm of specific variants that may cause symptoms beyond those observed from the original strain.
Graphen’s early warning at the end of September 2020 on the Intl. Symposium on COVID was about 2.5 months ahead of UK government’s warning. Our binding energy prediction model is proven better than the current leading state-of-the-art in predicting ΔΔG.
Learn more about our work here.
