Scottsdale, AZ – June 1, 2019 –Systems Oncology, an AI-based biopharmaceutical company, is presenting a poster at the American Society of Clinical Oncology (ASCO) Annual Meeting in Chicago on June 1st, 2019. The presentation will highlight AI research using the ExpansiveAI platform, the company’s proprietary knowledge network that mines trillions of data connections to generate and test massive numbers of hypotheses using cognitive computing and machine learning to rapidly discover novel cancer drug targets. The method leverages one of the largest hypergraph knowledge-bases to efficiently extract many dimensions of target network features. These hypergraph network features were used to train an artificial neural network, which then classified more than 400 million gene pairs, and identified several hundred candidate synthetic lethal targets. Further bioinformatic filtering and in-silico validation led to the prioritization of specific drug targets which are predicted to create a contextually defined vulnerability in molecularly defined cancers. In other words, targets that will only be vulnerable in cancer cells that harbor a specific genetic change. This work led to the launch of new therapeutic programs at Systems Oncology that promise to be much more selectively effective and safer than traditional therapeutic approaches.
“We have had a long-standing interest in discovering new synthetic lethal targets in cancer, but the full universe of synthetic lethal gene pair possibilities is far beyond what the human mind can process. Our thesis is that applying this unique big data and AI driven approach could help us rapidly explore a much more expansive universe of therapeutic opportunities. The results of this research, which was a collaboration between Systems Oncology and Systems Imagination, are far beyond what could ever be explored through human cognition alone. The results were truly transformative for our ability to discover new cancer biology, so we are excited to share this new approach with the cancer research community.” said Dr. Spyro Mousses, Chief Executive Officer of Systems Oncology. “What is truly incredible here is the scale of this kind of research and how empowering it is for finding novel insights into cancer biology. The cognitive computing platform explored massive volumes of empirical data, but then applied that learning to autonomously mine an unprecedented universe of hypotheses, leading to the identification of truly novel synthetic lethal targets that we’ve never seen before.”
Abstract and the full poster are available online.
Title: Application of artificial intelligence to predict a new class of novel synthetic lethal targets
Session: Developmental Immunotherapy and Tumor Immunobiology
Board #: 242
Time: Saturday June 1st ,2019, 8AM-11AM
Location: Hall A
About Systems Oncology
Systems Oncology (SO) combines multi-scalar systems modeling with machine learning and big-data to uncover breakthrough insights into cancer biology and expose new therapeutic vulnerabilities. These discoveries are then used to better develop truly novel therapeutic agents. SO has the experience and knowhow needed to intelligently translate scientific insights into smarter clinical development strategies and faster registration paths. It then partners with leading pharmaceutical companies to accelerate the development of breakthrough cancer treatments. Fundamental to the SO approach is a next-generation AI platform for translating scientific insights into innovative cancer therapies with unprecedented speed, scale, and precision.