Evaluations for the Computational Methods for Spatially Realistic Microphysiological Simulations Workshop

April 2015

 

Participants were asked to evaulate the workshop in the following areas. Numeric scales ranged from 1 -5, with one being the lowest satisfaction and 5 being the highest.

The Cell Modeling workshop was very valuable for my current/future research

1 2 3 4 5 Average score
- - 1 3 12 4.7

I would recommend the Cell Modeling Workshop to my colleagues

1 2 3 4 5 Average score
- - - 3 13 4.8

The instructors and teaching assistants were well prepared.

1 2 3 4 5 Average score
- - 2 7 7 4.3

The instructors and teaching assistants were helpful and assisted me with my problems and questions.

1 2 3 4 5 Average score
- - 1 3 12 4.7

The online tutorials were useful (www.mcell.org/tutorials).

1 2 3 4 5 Average score
- - - 9 7 4.4

Which aspect of the workshop was most useful to you?

  • Learning getting scientific figure from the biological study
  • The combination of theory and tutorials. I think on their own, each would be useful, but not nearly as useful.
  • Exposure to recent developments especially CellBlender. Opportunity to interact with developers. New modeling possibilities even though I may not use them.
  • For me (a biologist with little computational/modeling experience) it the most useful were (tie): -very basic aspects (e.g. programming theory/ nomenclature/ logic/ workflow) -afternoons dedicated to small group (or ""1-on-1"") work with workshop instructors and TAs.
  • I can get clear ideas of cell modeling and the latest progress.
  • Learning about MCell and CellBlender
  • The hands-on time and discussions with the instructors.
  • Most of the topics were knew to me and as a result I learned a lot!
  • Hands on lab exercises were most useful
  • MCell and CellBlender
  • Introductory seminars for MCell and BioNetGen (morning sessions), Introductory tutorials, Undirected hands-on time with programs"
  • MCell internals, CellBlender material, discussions with the developers.
  • Diffusion simulation methods and what we can get out of it.
  • The help of each assistant was exceptional good. There was always someone who could help out at every topic you can imagine.
  • The hands on tutorials in the computer room were most useful. However, the instructors moved too quickly, and there was little to no time to work on our own research projects - perhaps adding an optional forth day for apply the tools to personal research projects would be helpful. The instructors were very good at helping anyone out who had fallen behind, but there were typically not enough instructors to go around.
  • The first day was most useful to me, both talks and labs.

 

What workshop topics would you like to see emphasized more in the future?

  • More time on the specific package(s) that I will be using - could perhaps be specified via a pre-meeting phone call/ Skype session (to help the instructors/TAs understand the detailed goals of individual participants(?))
  • Explanation of details of modeling by using any realistic model.
  • None, the workshop was pretty close to perfect.
  • No idea!
  • More scheduled time for working on independent projects with assistance from staff.
  • Basic principles upon which the MCell etc. is based. There is a great danger in using these tools "blindly", and the danger increases as they become more user-friendly and attract wider user base.
  • A smaller research example of diffusion simulation.
  • I would recommend even more time for practical courses to train the software on your own.
  • I would spend less time lecturing about the mathematics behind the tools and more time lecturing about practical problems/questions that may arise when using the tools.
  • Larger focus on MCell. More lab exercises with specific tasks to accomplish.

Other comments

  • Be more realistic about the timing of lectures and tutorials. It ended up working, but it seemed a little chaotic at times.
  • Very nice atmosphere. Everyone very interested in topics and paying full attention.
  • Excellent and valuable workshop.
  • Seminars relating to recent thesis work were not useful and related demos did not work well on personal laptops.
  • First and third days were much better than the second day, especially the afternoon tutorial.
  • Overall, I liked Cell Modeling workshop very much. I was not too happy with the second afternoon. Instructors were going through material too fast.

The presentation was clear and well structured.

1 2 3 4 5 Average score
- - - 2 14 4.9

The lecture taught me the basics of building MCell Models

1 2 3 4 5 Average score
- - 1 3 12 4.7

I expect that MCell modeling will be important to my current or planned research projects.

1 2 3 4 5 Average score
- - 2 3 11 4.6

Comments or Suggestions

  • See earlier comment on parallel sessions.
  • 1. One day is not enough for this material.

    2. MDL deserves more time: There may be exceptions (very simple models) but it is unlikely that one could (or even should) get by entirely without writing or at least thorough understanding of MDL. When the model geometries have to be numerically accurate, CellBlender cannot easily create them anyway. So it's far more important that CellBlender reads and correctly displays every bit of legal MDL than that it writes MDL as well. Except for the simplest cases, it's also more time-consuming and error-prone to manage a collection of dialog boxes than to edit an MDL file.

The presentation was clear and well structured.

1 2 3 4 5 Average score
- - - 3 13 4.8

The lecture taught me the basics of MCell methods for diffusion and reaction.

1 2 3 4 5 Average score
- - 1 3 13 4.7

Comments

  • See earlier comment on parallel sessions.
  • More time was needed, especially to explain the algorithms behind more complicated reactions.

Ease of use of CellBlender

1 2 3 4 5 Average score
- - 7 5 4 3.8

I expect that CellBlender will be important to my current or planned research projects.

1 2 3 4 5 Average score
- - 3 3 10 4.4

In your opinion, what aspect of CellBlender needs the most improvement (e.g. model design, running simulations, analysis)?

  • Analysis: changing scales in graph outputs
  • Not possible to say without more familiarity
  • Drop down menus for some elements
  • Analysis
  • None, I think most of the efforts at improvement that were discussed are well targeted.
  • Model design
  • Model design, access to error messages
  • It should read and display any legal MDL!
  • Expand the library of cell shapes and organelles and make the export to SBML easier.

Are any features missing from CellBlender or MCell which are crucial for your workflow? Please elaborate.

  • In Window, some menus do not appear. For example, matplotlib for plot does not appear at my laptop
  • Not possible to say without more familiarity
  • Various methods for dynamics
  • It seems work on most of the missing features is currently underway.
  • n/a
  • Concentration-based simulation
  • Running multiple instances of MCell through CellBlender with different parameters

Comments or suggestions

Clarity of Presentation

1 2 3 4 5 Average score
- - 3 7 6 4.2

How much you learned

1 2 3 4 5 Average score
- - 4 7 5 4.1

Comments

  • Hands on Rulebender need much more time for the beginner like me to learn fully. I think it would start 11 AM at least.
  • Talk could have been a bit tighter in organization and timing but still very interesting.
  • The only confusions came when Jim's very well-organized (!) lecture on the basics of RBM got off onto rather lengthy (1-5 minutes) tangents. They got reeled in eventually. Shorter tangents (<1 min) weren't a problem
  • It's a difficult design task but the scripting language seemed a bit "young and awkward" at this point (compared to, say, MDL).
  • I didn't use RuleBender untill now so I don't know if this tool will be used in my further research. The tool itself seemed very powerful and useful!

Clarity of presentation

1 2 3 4 5 Average score
- - 3 8 5 4.1

How much you learned

1 2 3 4 5 Average score
1 2 5 5 4 3.6

Pace of the tutorial

Too fast About right
5 7

Ease of use of RuleBender

1 2 3 4 5 Average score
- 1 6 5 4 3.8

Usefulness of RuleBender to your current or planned research projects.

1 2 3 4 5 Average score
- 3 6 3 4 3.5

Comments

  • It was too fast. In addition to that, assigned time is so short that we (maybe) could follow them easily.
  • Demonstration was buggy and too fast-paced for trouble shooting.
  • Without prior knowledge, it's impossible to listen to a talk, watch the presentation, and do the tasks on your computer, all at the same time. The tutorial could be done similarly to the first day tutorials.

Are there any additional features that you would like to have in RuleBender or in the interface with CellBlender/MCell? Please elaborate.

  • For example, the lecturer gives a simple biochemical reaction and the students apply the RuleBender's rule to it
  • Allow binding interactions between molecules in two different membrane compartments, e.g. binding of ligand and receptors on two opposing cell surfaces.

Please add any other suggestions you have for improving Day 2 of the workshop.

  • Of the three days, this is most important. As I said just before, the time for Hands-on needs much more than this year.
  • I think it would help if tutorials were debugged more and were less ambitious.
  • More time for undirected hands-on work would be useful, particularly if staff were available for assistance.
  • From my (perhaps narrow) point of view, the RuleBender concept is relatively simple, the rest being the mechanics of doing it. Since the concepts and underlying algorithms are the most important things to learn, I'd suggest shortening this section to 1/2 day and use the time for MCell algorithms, which are conceptually more difficult.
  • Try afternoon sessions on naive computer-non-savvy audience.

Copyright © 2020 National Center for Multiscale Modeling of Biological Systems. All Rights Reserved.