Koga & Koga, as seen in PubMed.

Koga & Koga refers to ""Principles for designing ideal protein structures”, a paper published in 2012 in Nature by Nobuyasu Koga, Rie Tatsumi-Koga, and other authors.

Koga & Koga describes promising secondary structure patterns for designing new proteins. The patterns are based on analysis naturally occurring proteins and large-scale simulations of artificial proteins by Rosetta@home and its network of volunteers.

The patterns are the basis of the shapes seen in the blueprint tool in Foldit. They're also closely related to the ideal loops condition found in most recent Foldit design puzzles.

See Design structures for a discussion of creating the Koga & Koga patterns in Foldit. The Koga & Koga technical supplement explains the technical terminology that's key to the rules identified in Koga & Koga.

See the complete article on PubMed.


The technical language in Koga & Koga can be translated into terms more familiar to Foldit players. Here are some examples.

"Unique ordered structures" refers to the idea that a given protein, identified as a specific sequence of amino acids, folds up into a single shape. This idea is closely related to Anfinsen's dogma.

"Kinked α-helices" -- an α-helix is just helix in Foldit. It's easy to create an "ideal" helix in Foldit, but helixes (or helices for purists) are often bent or kinked in real proteins.

"Bulged β-strands" -- a β-strand is a sheet Foldit. Technically, you need two or more strands bonded together to make a true sheet. Again, Foldit can make ideal sheets, which have a slight curve, but real-life sheets can be a little messy.

"Buried polar groups" -- some amino acids are "polar", meaning they can form hydrogen bonds. These amino acids are generally more likely to be found on the surface of a protein, and are less likely to be buried in the core. But buried polars occur in real proteins.

"Protein tertiary motifs" -- the shapes of proteins. A motif is a shape that appears in different variations in different proteins (or in part of the same protein). Similar to a musical motif.

"Funnel-shaped protein folding energy landscapes" -- the idea that as protein folds up, it follows a folding funnel, with the pointed neck of the funnel being the "native" shape of the protein. The Foldit scientists look for an funnel-shaped result when they're evaluating proteins designed by Foldit players. See Improvements in Foldit designs for a example of the graphs that generated when Rosetta@home evaluates Foldit designs.

"Negative design" -- the idea that a protein must not only fold up into a stable shape, but it must also avoid getting stuck in a non-native shape as it folds.

"Local interactions" -- chemical bonds that form between segments that are numerically nearby. Koga & Koga looks at these local interactions as a first step, on the theory that interactions between more distant segments form later as the protein folds. Since they're held together by strong peptide bonds, segments that are numerically nearby are guaranteed to be close to each other in space. As the protein folds up, more numerically distant segments can move closer in space, letting them form hydrogen bonds and other weaker types of bonds.

The Koga & Koga technical supplement delves into "chirality", "parallel", and "anti-parallel", which are key to understanding the rules that Koga & Koga identifies for low-level structures.

Community content is available under CC-BY-SA unless otherwise noted.