A point is that we actually do have a question in the final survey that reads "What are the Bioinformatics research areas of interest that you wish to pursue?" We could just combine the 2 questions together and get a better view.
I need help however in consolidating the answers to be on similar categories. Also, note that the later question was a multiple choice list, whereas the former is from a ticking box. Meaning that counts are not simply lummaple
[I think I would remove this figure all together!]
[The figure of students' interests may explain: 1) dropout rate, 2) students' mindset and expectations from the course (c.f. commercial bioinformatics courses in Sudan , 3) have students changed their interests by the end of the course?)]
For students interests after the course, we did ask a different kind of question, and gave them very different options, so their answers were:  Comparative Genomics,  Database resources, Genome annotation, Genome variation, Phylogenetics, Proteomics and Structural Bioinformatics, Sequence Alignment       
In either case, proteomics is interesting because it is a buzz word, and the potential for drug design, vaccine. we also have bias towards infections disease