When I bumped into my biochemistry professor, Dr. Kevin Ahern, on campus a few months ago, I had the pleasure of explaining how I actually get to use what I learned in his class. And at the Open Source Lab of all places. At the lab, I’ve had the opportunity to work on an open source project called the Protein Geometry Database (PGD), and my coursework as a food science major with fermentation science option – specifically, that course in biochemistry – has proven unexpectedly helpful when working on the PGD.

The Protein Geometry Database was originally created in 2008 by Dr. Donald Berkholz and Dr. P. Andrew Karplus, in conjunction with Peter Krenesky and John Davidson at the Open Source Lab. An extremely brief summary of the database: proteins consist of polypeptide chains, which themselves consist of amino acid residues. The dihedral angles between the residues and other characteristics of the bonds found within are influenced by the side chains of those residues and of their neighbors near and far. These characteristics are among the many attributes that are of interest to researchers like Drs. Berkholz and Karplus, both of Oregon State University’s Department of Biochemistry and Biophysics. Since that time, the PGD has evolved and matured to fit the needs of researchers both on- and off-campus, maintained primarily by student developers at the Open Source Lab. An example of this is the usage of PGD by cancer researchers at Fox Chase Cancer Center in Philadelphia to improve their understanding of structural details at the 0.1 angstrom level, which may assist them in identifying compounds of interest for cancer-fighting drugs.

All software development projects require two sets of skills: those pertaining to the various technologies (languages, frameworks and the like), and having a basic understanding of the concepts being modeled. The PGD is written in Python, and currently uses the Django framework to provide a web interface to a database containing protein sequences. Having experience with Django and Python is necessary, but not sufficient, because familiarity with protein structures is required to truly grok the PGD codebase.

One example from the PGD is Ramachandran plots. These plots help researchers identify secondary structures in proteins. Two of these structures – alpha helices and beta strands – were discovered by Linus Pauling OSU ‘22. Having covered these plots and their purpose in biochemistry class gave me a leg up on understanding the software used to generate the plots in the PGD.

Hold on tight, we’re about to get seriously technical.

The sequence of amino acid residues in polypeptide chains is the primary structure, while the interactions between nearby residues generates the secondary structure. Amino acids are the building blocks of proteins. They are composed of three groups surrounding a central carbon atom: an amine (-NH2) group, a carboxylic acid (-COOH) group, and a side chain. A peptide bond is formed between two amino acids when the amine group from one amino acid loses a hydrogen atom while the carboxylic group from another amino acid loses a hydroxide ion. The amino acid residues remain bonded while the hydrogen atom and hydroxide ion form a water molecule and are lost. The backbone of the polypeptide chain therefore consists of three atoms per residue: the nitrogen atom from the amine group, the central carbon atom, and the carbon atom from the carboxylic acid group. Bonds between these atoms have three dihedral angles associated with them: phi, psi, and omega. Phi and psi are heavily influenced by the side chains of the residues while omega is primarily influenced by the planarity of the peptide bond. For more information on the topic, download “Biochemistry Free & Easy” by Dr. Kevin Ahern and Dr. Indira Rajagopal at http://biochem.science.oregonstate.edu/biochemistry-free-and-easy. This was the textbook for my biochemistry class and is chock-full of Dr. Ahern’s poetry and songs as well as the sort of information one has come to expect from high-quality college textbooks.