Metagenome Annotation Using a Distributed Grid of Undergraduate Students

Community Page B ioinformatics is at the crossroads of different scientific disciplines, in particular biology, mathematics, and computer sciences. While this interdisciplinary aspect undoubtedly contributes to the subject's attractiveness for researchers and students, it implies the ability to master heterogeneous skills. For biology students, preconceptions lead some to believe that in silico approaches are for computer-savvy specialists only. Teaching bioinformatics not only implies helping students to overcome perceived obstacles, such as mastering biostatistics or computational tools. It also requires momentous efforts to effectively drive home the message that applying bioinformatics tools and interpreting their results is an eminently biological endeavor. With bioinformatics progressively entering core life sciences curricula, these challenges are being faced by an increasing number of universities. Luckily, since most bioinformatics resources are accessible online, almost every type of bioinformatics teaching can be done from a computer room equipped with broadband internet. Early on in their undergraduate studies, students can tackle bioinformatics questions that are at the forefront of current research, and even investigate problems that have not been addressed to date. Another advantage of in silico teaching is that processes are relatively fast and can be repeated over and over again at little cost, which encourages the learning process through " trial and error " iterations. However, easy data generation has its pitfalls: raw data flooding and, more perniciously, overconfidence in predictions. The current feeling that " turning data into knowledge " [1] is the major bottleneck in science in general, and particularly in postgenomic bioinformatics, is of immediate and extreme pedagogical importance. In this report, we present our experience using public cutting-edge genomic data, combined with a newly developed online environment for teaching bioinformatics at undergraduate level. The approach combines the excitement of novelty provided by " hot-off-the-sequencer, " as yet non-annotated metagenomics data, with a highly structured e-learning Web tool. We discuss its practical use in class together with assessments made by students. The teaching approach we present is the result of several years of experience in teaching bioinformatics at undergraduate level, during which we identified the following points as key ingredients for a successful teaching approach: Learning by doing: at undergraduate level, bioinformatics is best introduced by first-hand experience; theoretical considerations are easier to grasp once students are truly familiar with the tools. Learning through repetition: mastering bioinformatics tools requires using and reusing them in a wide range of situations. Confronted with " twilight …