Table summarizing the NCI/DOE Cellular Level Pilot team's aims and accomplishments during the first calendar year of JDACS4C. Text reads "AIM. Devlop reliable machine learning-based predictive models of drug response that enable the projection of screening results from among cell line, cell-line xenographfts, and PDX models. Increase the efficacy with which cancer drugs are selected for use with increasingly distinct patient populations, while accelerating the identification and evaluation of promising new cancer drugs. Extend herory and tools for QU in multiple spatial and temporal scales, advance development of a computational framework for integrating theory, simulation, and wet-lab experiments with UQ. Integrate UQ and optimal experimental design for assert Accomplishments. Established a database populated wiht NCI data sources (CANDLE release). Build hybrid modeling frameworks for shallow and deep learning version of problems. Applied multiple methods of feature selection for each agent, or class of agents, to generate compact molecular signatures that retain predictive performace- reduced less than 50,000 characteristic features to 50 or fewer while retaining an accuracy less than .97.Developed new methods for model abstraction and hypothesis formation.