The College Board has started its public-relations push for the coming changes to the AP biology exam, with a mostly favorable article in the New York Times: Rethinking Advanced Placement. The changes are in a good direction: less memorization of factoids, more long-response questions, some slight addition of math.
The changes don’t include one change that the International Society for Computational Biology would really like to see: an integration of bioinformatics tools into introductory biology courses. The draft of the proposed changes has no mention of bioinformatics and “comput*” finds only 3 instances on 63 pages. So the AP bio course is still planned to be 20th century biology, with essentially no mention of the big changes in method that have arisen in the past 2 decades. This is not really surprising as the process for making changes to the AP curriculum is glacially slow, as the the changes being rolled out in 2012 were started in 2002.
I looked a little at the curriculum to figure out where bioinformatics can be naturally added to enhance the learning of basic concepts.
One obvious place is under “1.A.4 Biological evolution is supported by evidence from many scientific disciplines”. Having students use bioinformatics tools to align DNA sequences from multiple organisms and build a phylogenetic tree of them would be useful enhancement, particularly if different genes were selected that had slightly different trees, so that students could learn some of the limitations of phylogenetic inference. (I think that many assume that the tree is very well known and that there are no questions about what tree is correct.)
Looking at conserved regions in the could also be useful for “1.B.1 Organisms share many conserved core processes and features that evolved and are widely distributed among organisms today.” Building phylogenetic trees from ribosomal RNA also helps support this concept.
“1.B.2 A phylogenetic tree and/or a cladogram is a graphical representation (model) of evolutionary history that can be tested.” is an obvious place to have students build their own trees. Having some explanations of branch length and bootstrap support would be useful, but may be beyond the scope of an AP class.
There are several different algorithms for generating phylogenetic trees. Explaining them in detail is probably beyond the scope of AP Biology, but a brief introduction to the flavors of different algorithms (neighbor-joining, maximum likelihood, and maximum parsimony, for example) might be reasonable. The distance measures used are based on (usually oversimplified) models of evolution, but explaining them might be tough for kids that haven’t had matrix multiplication yet.
It might also be worthwhile to explain when it is appropriate to do trees based on DNA and when to do them on protein sequences (basically, DNA-based trees are more sensitive to small divergences, and protein-based ones more accurate for large divergences). Examples of both could make a nice comparative study.
For “1.B.3 Non-eukaryotes can transfer genetic information laterally through the mechanisms of transformation, transduction and conjugation; most eukaryotes do not transfer information laterally”, is there an easy way to demonstrate lateral transfer between Archaea and Eubacteria using the UCSC microbial browser?
It shouldn’t be too hard to find examples of lateral transfer, but how do we convince the students that they were lateral transfer rather than deletion from a common ancestor? Perhaps the easiest cases are recent bacteria->archaea transfers where the GC content has not equilibrated to the background GC content of the receiving genome. Perhaps we could also look for unusual codon usage, though that is not a built-in track on the browser, so would require an additional tool. Still it would be a fairly easy python script to write, and could be distributed free on a website.
Other reasonable things to do are to use BLAST to find homologs for genes, to try to track transmission of HIV in public databases of HIV sequences, to isolate and sequence bacteriophages, and to look at protein structures with molecular viewing software (using jmol on proteopedia, for example). There are more ideas in the comments on my previous post about bioinformatics in high school.
Note: the sequencing of bacteriophages is currently a bit expensive for high schools, as a sequencing run sufficient for a class’s worth of bacteriophages costs about $2000. New equipment such as the Ion Torrent machines will bring that down to about $500 (mainly library prep), and it would be very good for colleges and universities to offer low-cost sequencing services to high schools.
The major revision happening in the AP Bio curriculum and the extra flexibility in lab exercises is a good opportunity for biologists and bioinformaticians to design new low-cost exercises that really use the tools of 21st century biology: particularly the large public databases of data that are waiting for people to discover things in them.
I’m on a task force that is supposed to be looking for ways to get bioinformatics into AP biology, but I get the impression that everyone gave up when they saw that it was too late to change the exams. The hard part is going to be coming up with well-documented exercises and labs that overwhelmed high-school teachers can pick up and use. Who is going to do the work of creating that stuff? It is not likely to come from current AP bio teachers, who are mostly overwhelmed trying to cram 4 years of biology into a 1-year course and don’t have time to breathe, much less try to shoehorn more material into their courses. (There are some exceptional teachers who have started to incorporate bioinformatics, of course, and we need to give them a forum for sharing their ideas and testing each others’ ideas.)
I’ve thought, on and off, about trying to design some exercises like this myself, but I’ve not had the time. I’ve created a lot of courses (added 3 new ones to our curriculum last year, and planning a major overhaul of another one this year), but these have been at senior and graduate levels, and I’m not sure I have the appropriate experience to scope exercises right for high schoolers and college freshmen. I’ve also had very little formal training in biology, so deciding what is important to teach at that level is difficult for me. I would have an easier time designing bioinformatics exercises to teach computer programming than to teach biology. I would be interested in working with biology teachers (at AP or college freshmen levels) in trying to come up with such exercises, though.