New Technique Allows Scientists to 'Listen' to Proteins

Researchers have come up with a bold new method for representing and understanding a protein's shape: translating it into music.

Ordinarily, to show the structure of proteins—found in every cell of every living thing—scientists create visual representations made of loops and folds and sheets.

However, these models are just that: one way of representing data. You can't take a snapshot of a protein, says Robert Bywater of London's Francis Crick Institute. Rather, visual models are created by measuring patterns that result when you shine a laser through a purified protein molecule.

Bywater has teamed up with Jonathan Middleton, a composer and musicologist at Eastern Washington University, to turn the shapes of the chemicals into sound. In a study published October 20 in the journal Heliyon, the duo turned the structure of three proteins into musical compositions that assign different pitches to the five common shapes found within each.

In the paper, 38 volunteers listened to portions of the compositions while looking at the corresponding structural picture. A majority of these people agreed that the sound correlated well with the visual model. "It allowed people to understand the structure better" by making it easier to follow and remember, Middleton says.

Taking note of these changes should help researchers be able to better compare and contrast different proteins, and better understand their evolution and function—which could lead to advances in many different fields of biology, Bywater says. Certain differences are easier to hear than to see, Middleton adds. For example, in complex proteins, certain shapes like loops or sheets may not be easy to see, because they are covered by or enmeshed in surrounding material. But sound can represent that shape without any problem, he says.

This method of turning visuals into sound, or "sonification," is innovative, says Guy Yachdav, a bioinformatics expert at the Technical University of Munich who wasn't involved in the research. It allows people to understand the structure sequentially, because sound is by its nature sequential. This may "help in pinpointing the different curvatures or pointing out…changes in the structure" that wouldn't otherwise be apparent, Yachdav says.

This paper is one of the more in-depth attempts at the sonification of a chemical structure, Middleton says The field that has been around for a couple decades but is still in its infancy, he says. Sonification has led to some advances in other areas, says Paul Vickers, a computer scientist at Northumbria University. In one case, work by researcher Robert Alexander on the musical sonification of solar wind data ended up being used by physicists to better understand some aspects of solar activity. In another, Vicker and colleagues turned computer traffic into sound, allowing them to spot patterns of malicious behavior and improve their security system, he says.

Middleton also created a pitch for each of the amino acids that make up proteins. He assigned these pitches based on their degree of attraction to—or repulsion from—water, which determines what parts of a protein bond with one another to form its 3-D shape.

The sequence of amino acids in a protein is relatively easy to work out. However, this linear structure doesn't tell you exactly how the string of amino acids will fold into a 3-D shape in the body. Figuring that out is one of the most important and difficult tasks in microbiology, Yachdav says.

Bywater says he envisions the technique being used more widely in research. "It does give you a feeling for structures" you wouldn't otherwise get, he says. "One of the things about music, it's also easy to remember."