Computational Biophysics Workshop - Pittsburgh, May 10-14, 2010

General Evaluation of the Theoretical and Computational Biophysics Workshop held at the Pittsburgh Supercomputing Center

May 10-14, 2010

Questionnaire: Pittsburgh Supercomputing Center
Analysis and report: David Brandon, TCB group, UIUC

The UIUC's Theoretical and Computational Biophysics Group (TCBG), the NIH Resource for Macromolecular Modeling and Bioinformatics is headed by Klaus Schulten, and CO-PIs Z. Luthey-Schulten,  L. Kale, E. Tajkhorshid, and A. Aksimentiev.  As part of its outreach, the Resource offers workshops to introduce and transfer its programs and technological solutions to the biomedical community.  The Resource participated in a one-week (May 10-14) workshop sponsored by the National Resource for Biomedical Supercomputing (NRBSC) at the Pittsburgh Supercomputing Center at Pittsburgh, Pennsylvania.  The NRBSC provided the classroom, and computer lab used for the workshop, as well as covering hotel, honorarium, travel, and printing costs; TCBG provided lectures and tutorials, and covered some printing costs.

The program of the workshop consisted of lectures and hands-on sessions. Workshop lectures were given by K. Schulten (UIUC), Z. Schulten (UIUC),  and E. Tajkhorshid (UIUC); teaching assistants helped participants during the hands-on tutorial sessions. 

After the workshop, participants were asked to complete an anonymous  workshop evaluation survey. The evaluation form asks about: topics and instructions, personal value, facilities and food service, and solicits overall comments and suggestions. Click here to see a sample evaluation form and response. A total of 22 evaluation forms were returned, providing an overall response rate of 73%. In terms of demographics, education levels of participants were high, including 11 doctorates, 17 graduate students, and one bachelor's degree.

The results of the general evaluation questionnaire are summarized below, within the following sections:

All responses in the following tables are reported in percentages, rows adding up to 100%. All respondents answered all items; the number of responses per item is presented next to each question. The scale used ran from a high satisfactory rating of "5" to a low least satisfactory rating of "1". Results referred to as 'highly rated' or 'positive' or 'majority' in the discussion below are based on adding the results for the values "5" and "4" as listed on the scale.

Table I. Topic and Instructions

Scale (+) 5 4 3 2 1 Scale (-) N/A N
The choice of subjects was well chosen 59% 41% poorly chosen 22
The instructors were excellent 73% 27% unsatisfactory 22
The methods of instruction were most appropriate 50% 41% 9% unsatisfactory 22
The prerequisite readings were very useful 18% 32% 41% of no benefit 9% 22
The lecture notes were very useful 45% 27% 18% of no benefit 9% 22
The hands-on examples were  very useful 77% 23% of no benefit 22

 

 All respondents provided high ratings of the the choice of subjects (100%), the instructors (100%), and the hands-on examples (100%). A majority also provided positive ratings for the methods of instruction (91%), and the lecture notes (69%). Half of the participants (50%) provided a positive rating for prerequisite readings (participants were asked to read the "Scalable molecular dynamics with NAMD" paper prior to the workshop.

Table II. Personal Value

  Scale (+) 5 4 3 2 1 Scale (-) N/A N
Informal conversations with other participants were very beneficial 32% 41% 23% 5% of no benefit 22
I gained new knowledge and insights definitely 64% 36% not at all 22
As a result of this program, I feel my research will improve greatly 50% 50% not at all 22
I would recommend this program to others highly 77% 18% 5% not at all 22

 

Regarding personal value, all participants indicated that they gained new knowledge and insights (100%), and that their research would improve as a result of the program (100%). Nearly all participants (95%) indicated they would recommend the program to others, and a majority of participants (73%) found that informal conversations with other participants were a positive experience.

Table III.  Facilities and Food Service

  Scale (+) 5 4 3 2 Poor-1 Scale (-) N/A N
The hotel facility was excellent 36% 9% 5% unsatisfactory 50% 22
The meeting rooms were excellent 68% 32% unsatisfactory 22
The computer room was excellent 68% 32% unsatisfactory 22
Food service and breaks were excellent 91% 9% unsatisfactory 22

 

All respondents agreed that the meeting rooms (100%), the computer room (100%), and the food service breaks (100%) were a positive aspect of the workshop. While half of the participants were local and didn't need to use the hotel, of those at the hotel a majority (91%) rated the hotel positively.

IV. Comments

The evaluation form also solicited written responses to open questions, on subjects such as meeting training needs, the value of topics, suggestions for future workshops, and how to advertise the workshop. The number of comments per question varied; three sample comments per question are listed below. Full comment results can be requested from the  organizers by emailing brandon@ks.uiuc.edu.

How did this workshop address your current training needs? (please be specific)

  • Helped in understanding the “black box”. Got new ideas regarding how to approach my project, and get more useful information out of my simulations.
  • I will use molecular dynamics modeling as a part of my research, and it was specially useful to be aware of the limitations and problems in the current implementation of molecular dynamics during the workshop lectures.
  • I had some knowledge of MD techniques, but felt I needed more hands-on experience, discussion of nuts and bolts of the sims, etc. - and got that from this workshop.

The topic most valuable to me was:

  • Free energy calculations - These provide another qualitative way of comparing structures and can lead to valuable insights. I plan to use what I learned about this at the workshop in the next few months.
  • Molecular dynamics, Great care was taken to explain not only the theory, but the practical aspects.
  • Setting up simulations of Membrane proteins. Because from scratch (i.e PDB file) the lecture and then the tutorial walks you though the whole process of setting up a simulation, and the the analysis in order to get valuable information.

The topic least valuable to me was:

  • I feel all subjects were valuable, even those that were most distantly related (e.g. membrane proteins) to my current field of work. This is so because each tutorial inevitably deepens the users’ hands-on experience with NAMD and VMD, and this familiarity with the potential of leading programs in the field aids in the planning and realization phases of any project in computational chemistry and (bio)physics
  • Sequence analysis. This has nothing to do with my research. (It was interesting, though.)
  • The lecture on doing simulations for carbon nanotubes. . Although I appreciate that this idea can be useful for simulating water channels through membrane proteins, I’m currently not involved in this kind of research, which is why it’s not that useful for me now. This could maybe change in the future, if I get involved in a project that
    involves studying membrane proteins.

In a future workshop of this type, I would like to have the following areas covered:

  • Residue-based CG
  • Common errors and trouble shooting using VMD and NAMD
  • Effective ways of analyzing and presenting simulation data. A bit more on steered molecular dynamics since its the tool I use.

How and where do you think we can most effectively advertise this type of workshop?

  • Computational Chemistry List (CCL), NAMD-List, NAMD/VMD Homepage
  • By emailing the registered people who download and use the software i.e when I downloaded VMD and namd, had to provide email details.
  • Via an email blitz!

Please recommend topics for future workshops or for a follow-up to this workshop.

  • Coarse grained simulations, free energy calculations.
  • A more detailed or “advanced” parameterization tutorial, including all required sub-steps (e.g. ab initio calculations) would be beneficial. Since most NAMD tutorials are based on the CHARMM force field, an advanced parameterization tutorial should provide parameters consistent with the CHARMM parameterization philosophy
  • I believe that the workshop already covered enough topics and in enough detail, so I can’t really recommend
    anything extra that can be done.