The INtegrated and Scalable PredictIon of REsistanace (INSPIRE) project aims to lay the foundations for the use of molecular simulation and machine learning to guide precision cancer therapy, in which therapy is tailored to provide maximum benefit to individual patients based on the genetic information about their particular cancer. In recent years, the development of targeted kinase inhibitors (which bind to proteins involved in signaling pathways that often control growth and proliferation that become dysregulated in many cancers) has changed the way many cancers are treated. Unfortunately, the development of resistance to these therapies limits the amount of time that patients can derive benefits from their treatment. Resistance to therapeutics is responsible for more than 90% of deaths in patients with metastatic cancer. The use of predictive simulation is vital to such an approach as the vast majority of clinically observed mutations are rare, essentially ensuring that it will be impossible that catalog-building alone will be sufficient for making therapeutic decisions.
The INSPIRE project was recently awarded a major allocation of super computer resources (80 million core hours on Titan, the world’s fourth fastest supercomputer, as of June 2017) through the US Department of Energy INCITE Supercomputing Award. The award will start from 1 January 2018 and will run for one year initially, renewable on approval for a second year. This novel proposal was led by Prof Peter V Coveney at UCL, and combines his teams work on binding affinity predictions into collaboration with Dr John Chodera from the Memorial Sloan Kettering Cancer Center in New York City; and with Rick Stevens at Argonne National Laboratory and the University of Chicago, who leads the Deep Learning Enabled Precision Medicine for Cancer project in collaboration with four DoE labs and the National Cancer Institute. The relevance to our own project efforts are compounded through the further collaboration of Prof Shantenu Jha from our Associate Partner Rutgers, and Dr Herman Van Vlijmen from our Core Partner Janssen. This is a project that will combine HPC with HPDA (high performance data analytics) using RADICAL Cybertools and we hope it will lead to an important breakthrough in computational biomedicine.
You can read more about this award, and the others that have been awarded for 2018 and previous years here