Abstract:

This paper describes a computer-assisted learning exercise whereby students studying structural biology use a short piece of protein sequence to identify that protein in online sequence databases. They then download a file for this protein from the three-dimensional structure database (PDB; www.rcsb.org ) and use computer graphics programs to view and analyze the structure. Sequences of proteins not in the PDB are then modelled by online homology modelling to create a structural model using Geno 3D (http://pbil.ibcp.fr/htm/index.php ) and some protein animations are created using a MultiGIF program (http://www.glycosciences.de/modeling/pdb2mgif/ ). The purpose of this exercise is to familiarise students with important online resources but also to encourage their curiosity and feeling of ownership of their learning. It is a process of “simulated discovery” in that the student does not know what they will find (as is the case in real research) but in fact the scenario is established so they will definitely find a protein and get to analyze its structure (hence “simulated”). Examples of student work will be included in this presentation to illustrate what can be achieved. The following provides more background and technical data for non-scientists.

Proteins are 1 of 4 principal categories of complex biomolecules in biology (others are fats, carbohydrates and nucleic acids). These are made by combining simpler molecules, providing a unifying theme to biology at the molecular level which is pedagocically useful. Nucleic acids and proteins are “informational” molecules i.e. their properties are determined by their building block sequence. Most of what happens in Biology is mediated ultimately by proteins which are, for example, responsible for structure (e.g. collagen, the main protein of skin), catalysis (e.g. enzymes), defence (e.g. antibodies), movement (e.g. the proteins responsible for muscle contraction; actin and myosin), storage (e.g. plant seed storage proteins), transport (e.g. haemoglobin transports oxygen from lungs to muscles) and control (e.g. protein hormones like insulin). All proteins in biology are composed of chemical building blocks, the amino acids, (of which there are 20 common ones) arranged in a linear sequence (primary structure). Each protein has a unique sequence or "primary structure" which is ultimately determined by a nucleic acid sequence (usually deoxyribonucleic acid or DNA) made up of a sequence of a different type of building block (of which there are four main types called nucleotides) encoded in the genetic material or genome. At its simplest, each protein has a gene encoding it. The human genome (sum of our genes) is now completely sequenced and is held in sequence archives which are web-accessible and searchable. Each amino acid in the protein "language" is determined by the sequence of three nucleotides (i.e. the amino acid methionine, M, is encoded by ATG). This relationship is defined by the universal genetic code which allows us to translate nucleotide sequences into amino acid sequences. The ease of translation and the linear nature of sequences make storing, reading and analyzing both nucleotide and amino acid sequences by computers completely feasible. The computer-assisted learning exercise described here takes advantage of this facility to find and read and compare protein sequences in silico from genome sequences but also shows how proteins can be appreciated in three dimensions.