Back to MAIN AGENDA
Day 1: Wednesday March 22, 2017
Moderators: Jennifer Couch (NCI)
Michele Grimm (NSF)
Barbara Sorkin (NIH/OD)
Speakers :
- 8:30 - 8:50 am: Andrew McCulloch
- 8:50 - 9:10 am: Ahmet Erdemir
- 9:10 - 9:30 am: Cecile Viboud
- 9:30 - 9:40 am: panel discussion
The objective of this session is to set the stage for the remainder of the meeting both from a historical perspective on the evolution of multi-scale modeling over the last several years and the impact of IMAG on that evolution coupled with a forward looking perspective: where are we now and what are the most exciting and/or pressing big challenges to be addressed over the next few years?
The specific charge to the speakers and for discussion are:
- to discuss their (current) work in multi-scale mathematical/computational modeling as an example of progress, success, and remaining challenges in multi-scale modeling;
- to address how progress in math/computational modeling over the last decade (especially IMAG and/or MSM contributions) has contributed to advancing their work;
- to address likely or most exciting future directions for their research or sub-discipline and how IMAG might evolve and adapt to meet the needs of emerging multi-scale modeling challenges.
Comment
How do limits on
For Dr. Viboud: Could the
For Dr. Viboud: Could the small city hubs be associated with the presence of nearby military bases?
For Dr. Viboud, part II: As
For Dr. Viboud, part II: As in, the military bases are likely hubs of transport.
When dealing with large
When dealing with large volumes of data (which is becoming more and more common), we often have to use advanced techniques to determine what effects our models are most sensitive to. When faced with many types of principal component analysis, t-SNE, etc., how do you decide the most appropriate method to perform analysis on your data?
What role may be played by
What role may be played by Virtual Cell and other modeling environments - particularly to distribute your "finished" programs.
How do limits on computational capabilities of available tools most impact progress in multiscale modeling? Where would computational acceleration most benefit progress?