Computer Vision News - November 2018

18 Challenge Computer Vision News With Tero Aittokallio DREAM has launched the IDG-DREAM Drug Kinase Binding Prediction Challenge . The challenge is currently open and it will run until December 2018. Winners will present their winning model at the DREAM 2019 Conference . The goal of mapping the bioactivity of the full space of compound-target interactions is probably unattainable, even using the most powerful hardware and software technologies: simply put, the size of the chemical universe is excessively large. However, it is possible to prioritize most potent interactions for further experimental evaluation. Due to their clinical importance, the organizers of this IDG-DREAM challenge have decided to focus on kinase inhibitors . The objective is to further extend the druggability of the human kinome space . This is to be achieved by evaluating the practical power of statistical and machine learning models to predict drug-protein binding affinity and allow to prioritize the most powerful interactions. This research requires a multi- disciplinary approach testing the boundaries and pushing the limits of both machine learning and drug discovery. The project is consequent to a 2017 study by then PhD student Anna Cichonska et al.: Computational- experimental approach to drug-target interaction mapping: A case study on kinase inhibitors (Anna has successfully defended her dissertation only a few weeks ago). The IDG-DREAM Drug Kinase Binding Prediction Challenge is launched to benchmark machine learning models in the effort to accelerate mapping of drug-target space by prioritizing most potent compound-target interactions for further experimental evaluation. Challenge Visualisationof the binding between tivozanib and ABL1 Sage Bionetworks