Building Bridges
Science Education at Bryn Mawr College

Second Annual Tri-College Math/Science Teaching Symposium

Discussion of "Bioinformatics: A HHMI Faculty Development Course in Computer Applications"

Moderator: Susanne Amador Kane, Department of Physics, Haverford College

Minutes taken by:Ali Erkan, Computer Science Program, Swarthmore College

Building Bridges Home

Teaching Matters

SCIENCE TEACHING IN THE TRI-COLLEGE COMMUNITY

Attendees in discussion after Suzanne's presentations: Suzanne Amador Kane (Haverford), Victor Donnay
(Bryn Mawr), Suzan White (Bryn Mawr), Elizabeth Vallen (Swarthmore), Steve Maurer (Swarthmore), Thomas Hunter (Swarthmore), Ali Erkan (Swarthmore)

• The web page for the faculty development course on bioinformatics is here

 

• The web page of National Center for Biotechnology Information is also worth visiting
  The teaching resources and tutorials at this site are useful. Here, most of the
  information is geared towards preparing a graduate level course.
  

Major points of the discussion:

• When people from different disciplines work together for learning bioinformatics related topics, two different approaches seem to work well. For people with a math background, it is easier to start with simple problems that can be mathematically analyzed. These problems can then be incrementally made more complex. For people with a biology background, it is easier to start with a real problem and then step by step go down to the mathematical foundations.
  
  
• In some cases, a "just in time" learning seems to pay off. In other words, instead of learning a large volume of prerequisites before dealing with any of the real/interesting problems, it is more productive to start with simple versions of the problems, even working through them by hand (e.g. a small BLAST case). This makes itpossible to spend the right amount of time on new topics.
• Visualization is an important element of learning this material. One option is to develop Java applets to help students comprehend the solutions rather than understand only on the abstract plane. Before committing any resources to such efforts, it is important to see if such applets have already been developed elsewhere to prevent "reinventing the wheel".
  
• In research communities and the industry, members of bioinformatics groups do not hold advanced degrees specifically in bioinformatics. Instead, these groups are typically made up of members who are experts in well established areas (e.g. molecular biology, parallel numerical computations, etc). Consistent with this way of doing things, trying to create a special bioinformatics major may not be the right approach at this point. Given the importance of bioinformatics, incorporating issues from this field into existing courses seem to be
  very useful for our students. Some possibilities:
  
  
  • (i) Elements in chem/phys labs; e.g., finding the sequence of a protein the students have already studied;
  • (ii) Data structure, algorithm, and concurrency issues in computer science courses with respect to the required computations in bioinformatics problems;
  • (iii) Case studies in probability and statistic courses.
    
    
• One of the unexpected and useful outcomes of this course was the observation of how different people use different methods of analyzing data (GIS example). Learning each others' ways is useful for all of us. Another one of the unexpected and useful outcomes of this course was that each one of the involved faculty was "reminded" of what it feels like to be a student. In terms of being better instructors in this and other fields we need to periodically refreshen our knowledge of the students' perspective.