Translating Nature into Medicine Using Machine Learning, Metabolomics, and High-Throughput Pharmacology

At Enveda, we are building the largest integrated dataset of plant chemistry in the world, tailor-made for drug discovery. With metabolomics analyses of thousands of plants linked to a knowledge graph (BIOEDGE) integrating humanity’s collective knowledge about plants and our internal high-throughput bioassay data, we are uniquely positioned to scale the discovery of first-in-class small molecules for complex human diseases. Identifying and isolating active compounds from plants has historically been a slow, iterative and labor-intensive process. While our algorithms are quickly identifying active features and chemical structures from plants, our bioactivity assays must scale at a similar pace. Over the past few months we have scaled our bioactivity assays by an order of magnitude by leveraging the Echo 655, while simultaneously improving assay quality. Rapidly miniaturizing our assays to 1536-well plates and leveraging the Echo has allowed us to screen our growing library across numerous assays within a few months. With every experiment, our dataset grows, and our platform improves. We continuously learn something about both our unique library and disease biology that we reintroduce into our dataset to fine-tune our algorithms. We are grateful to partner with Beckman Coulter Life Sciences to explore the millions of phytochemicals that exist in nature to rapidly identify active molecules to treat human disease.