Deep Apple Therapeutics discovers the therapeutics for high-value targets through virtual screening of artificial intelligence (AI)-generated virtual libraries.
With a powerful discovery engine that combines ensemble cryo-EM, deep learning, and molecular docking screens of ultra-large libraries, Deep Apple can go from target identification to lead optimization in less than 12 months – a fraction of the industry standard time – and can pursue biological target signaling inaccessible to conventional discovery approaches.
“ATP created Deep Apple to revolutionize drug discovery in terms of speed, cost, and effectiveness,” said Spiros Liras, founding CEO of Deep Apple and a Venture Partner at ATP.
“We brought together unique capabilities from our founders to build a true deep learning discovery engine that stands apart from the pack of AI-driven approaches to protein structure elucidation and drug discovery.
“Machine-learning enabled processing of cryo-EM data allows us to reveal biologically relevant conformations in the context of interactions with signaling partners – transient binding pockets that may be missed by static models and empirical screening methods. And our virtual large-scale docking enables us to quickly home in on the right drug for the right target.”
The discovery engine is particularly well-suited to expedited hit-finding against integral membrane proteins and is broadly applicable across disease areas.
The company currently is advancing multiple programs focused on GPCR modulators, a proven target class with applications in metabolic disorders, inflammation, immunology, and endocrine diseases.
Deep Apple’s drug discovery engine builds upon leading expertise and technologies from its academic co-founders Georgios Skiniotis, a world leader in cryo-EM and GPCR structural biology and Brian Shoichet, a pioneer of virtual screening as well as John Irwin, the computational library authority who created the widely used ZINC free virtual library of more than 10 billion synthesizable compounds.
“Deep Apple exclusively uses virtual screening for hit identification, and we have achieved high-quality hits against difficult-to-drug targets at an extremely fast pace,” said Paul Da Silva Jardine, chief scientific officer at Deep Apple and a Venture Partner at ATP.
“Since we commenced operations last year, we have initiated multiple GPCR programs, including non-peptide/non-GLP-1 programs in obesity and weight management, as well as promising programs in inflammation and inflammatory disorders. And with the versatility of our discovery platform, GPCRs are only the tip of the iceberg.”
Deep Apple conducts in silico screening of billions of synthesizable compounds against orthosteric and allosteric binding sites in mere hours, and then computationally generates vast project-specific virtual libraries to discover proprietary chemotypes with desirable dockable and druggable properties. Wet-lab interrogations of the chosen virtual compounds feed back into the company’s deep learning models to continually improve predictive performance.
Seth Harrison, ATP Founder and managing partner, said: “ATP founds and builds companies that bring together critical technologies and efficient research plans for translation,” said. “In my more than three decades of investing in biotech startups, particularly in early-stage drug discovery platform technology, I have never seen a company to move as rapidly as Deep Apple has, from chemical biological idea to development candidate, for high-value, difficult-to-address targets.”
