What is it about?
Proteins are essential biological molecules. They are the building blocks of cells and control many life processes. Each protein is a chain, consisting of smaller subunits called amino acids. Depending on its amino acid composition and length, each protein can fold into a special 3D structure. The role and function of a protein depends on its 3D structure. Hence, protein structure has received immense attention from researchers. In this study, the authors developed an artificial intelligence (AI) tool to design and analyze proteins. They translated known proteins into pieces of music in which different pitches represented different amino acids and note length and volume represented protein length and 3D structure. Their AI tool could “learn” known music pieces (i.e., known proteins) and produce new music pieces (i.e., novel proteins). Hence, their tool could design artificial proteins that do not exist yet. Moreover, it could predict how changes in amino acids and 3D structure could affect the function of a protein.
Featured Image
Photo by Marius Masalar on Unsplash
Why is it important?
Proteins have wide applications in fields such as medicine and engineering. Moreover, disruptions in protein structure can cause diseases. Hence, tools for analyzing proteins and designing new ones are the need of the hour. The AI tool developed by the authors can be used to design new proteins with exciting applications. Moreover, it can help us gain insights into how changes in amino acids and 3D structure affect protein function. This can help tackle diseases in the future. KEY TAKEAWAY: By converting protein structures into music pieces and processing them on an AI system, the authors developed a robust tool for protein design and analysis. This tool can be applied in research on diseases caused by protein dysfunction. Notably, it can be used to design new proteins with various applications.
Perspectives
With the new method, we can listen to the music of life, understand the fundamental physical concepts of protein folding in novel ways, and exploit the power of AI to develop new molecular designs. Applications may include the development of de novo proteins and drugs by designing protein sequences. Audible coding of matter can also enable material design by manipulating sound, detecting mutations, and offering a way to reach out to broader communities to explain the physics of proteins. The nanomechanical structure of matter, reflected in an oscillatory framework, presents a new palette for sound generation, and can complement or support human creativity.
Markus Buehler
Massachusetts Institute of Technology
Read the Original
This page is a summary of: Sonification based de novo protein design using artificial intelligence, structure prediction, and analysis using molecular modeling, APL Bioengineering, March 2020, American Institute of Physics,
DOI: 10.1063/1.5133026.
You can read the full text:
Resources
Contributors
The following have contributed to this page
