researchers often fail to allocate enough upfront time for surveying a problem domain and getting a sense for the current state of the field as a whole. what are the most pressing questions in the community at the moment? what are the latest publicly available data sets? how to ensure that one's work will, in fact, uniquely contribute to the literature and be useful to the community as a whole?
rather than organize more meetings where researchers present their particular work in granular detail, i believe there is more to be gained from hearing these same scientists present/pitch POTENTIAL projects to an audience. this sort of collaborative and open discussion would expose younger scientists to fresh ideas far more efficiently than simply reading papers & attending conventional talks.
so let's get the ball rolling with some interesting open problems in computational cancer research. the ideal format would be a 20-30 minute pitch, followed by a lengthy group debate about the "optimal" way to leverage current data/methods to actually solve the proposed problem.