We proudly announce our global Artificial Intelligence (AI) collaboration with Atomwise, resulting in a new publication in the journal Scientific Reports.
The paper, “AI is a viable alternative to high throughput screening: a 318 target study” comes from the AI computation drug design company, Atomwise in San Francisco. cc-TDI was among 282 Institutions across the world to work with Atomwise to use the AtomNet artificial intelligence model, which is a graph convolution network architecture, to predict compounds that dock to a given disease-related protein. cc-TDI previously published results of their target in the British Journal of Cancer related to the adolescent and young adult cancer, clear cell sarcoma in the article Functional genomics of human clear cell sarcoma: genomic, transcriptomic and chemical biology landscape for clear cell sarcoma.
“The breadth of this collaboration across so many types of proteins (enzymes, nuclear receptors, transcription factors, DNA/RNA binding proteins, ion channels, transporters and GPCRs) and the observation that these computational-derived compounds bound and affected function of these proteins is remarkable” explains cc-TDI’s Scientific Director, Charles Keller.
cc-TDI has done similar work with the IBM-created World Community Grid and related pilot studies with Microsoft Azure HPC, lending confidence that the first stage of drug development can be rapid, inexpensive, and tractable for rare childhood cancers.
