Imagine designing a completely new protein with unique beneficial properties simply by composing a melody. This revolutionary concept is now becoming reality thanks to groundbreaking artificial intelligence technology developed at MIT.
In an extraordinary fusion of scientific innovation and artistic expression, researchers at the Massachusetts Institute of Technology have created an advanced system that transforms the intricate molecular structures of proteins—the fundamental building blocks of all life forms—into captivating musical compositions. This cutting-edge technology then reverses the process, allowing scientists to modify these musical sequences and convert them back into entirely new proteins that have never existed in nature.
While not quite as simple as humming a new protein into existence, this innovative system comes remarkably close. It establishes a systematic methodology for converting a protein's amino acid sequence into a musical sequence, utilizing the physical properties of molecules to determine the resulting sounds. Although these sounds are transposed to fall within the human audible range, the tones and their relationships are directly based on the actual vibrational frequencies of each amino acid molecule, calculated using advanced quantum chemistry theories.
This pioneering system was developed by Markus Buehler, the McAfee Professor of Engineering and head of MIT's Department of Civil and Environmental Engineering, in collaboration with postdoctoral researcher Chi Hua Yu and additional team members. As detailed in the journal ACS Nano, this revolutionary technology translates the 20 types of amino acids—the fundamental building blocks that combine in chains to form all proteins—into a distinctive 20-tone musical scale. Consequently, any protein's extended sequence of amino acids becomes a unique sequence of musical notes.
Although this scale may sound unfamiliar to those accustomed to Western musical traditions, listeners can quickly recognize the relationships and differences after becoming familiar with these novel sounds. Buehler notes that after listening to the resulting melodies, he can now identify specific amino acid sequences that correspond to proteins with particular structural functions. "That's a beta sheet," he might observe, or "that's an alpha helix."
Decoding the Complex Language of Proteins
The fundamental concept, Buehler explains, is to gain deeper insights into understanding proteins and their immense diversity of variations. Proteins constitute the structural material of skin, bone, and muscle, while also serving as enzymes, signaling chemicals, molecular switches, and numerous other functional components that comprise the machinery of all living organisms. However, their structures—including the way they fold into shapes that often determine their functions—are extraordinarily complex. "They have their own language, and we don't understand how it works," he explains. "We don't know what makes a silk protein a silk protein or what patterns reflect the functions found in an enzyme. We don't know the code."
By translating this complex language into a different form that humans are naturally attuned to—and which allows different aspects of information to be encoded in various dimensions such as pitch, volume, and duration—Buehler and his research team aim to uncover new insights into the relationships and differences between various protein families and their variations. This approach serves as a powerful method for exploring the countless possible modifications of protein structure and function. Similar to music, the structure of proteins is hierarchical, featuring different levels of structure at varying scales of length or time.
This innovative method translates amino acid sequences of proteins into these sequences of percussive and rhythmic sounds. Courtesy of Markus Buehler.
The research team then employed an advanced artificial intelligence system to analyze the extensive catalog of melodies produced by a diverse array of different proteins. They had the AI system introduce subtle modifications in the musical sequences or generate entirely new compositions, then translated these sounds back into proteins corresponding to the modified or newly designed versions. Through this process, they successfully created variations of existing proteins—for instance, one found in spider silk, one of nature's strongest materials—thus producing new proteins unlike any created through natural evolution.
The percussive, rhythmic, and musical sounds heard here are generated entirely from amino acid sequences. Courtesy of Markus Buehler.
Although the researchers themselves may not comprehend the underlying rules, "the AI has learned the language of how proteins are designed," and it can encode this knowledge to create variations of existing versions or entirely new protein designs, Buehler explains. Given that there exist "trillions and trillions" of potential combinations, he notes that when it comes to creating new proteins, "you wouldn't be able to do it from scratch, but that's exactly what the AI can accomplish."
"Composing" Revolutionary New Proteins
By utilizing such a system, he explains that training the AI system with a dataset for a specific class of proteins might require several days, but it can subsequently generate a design for a new variant within microseconds. "No other method comes close," he asserts. "The limitation is that the model doesn't reveal what's truly happening internally. We simply know it functions effectively."
This method of encoding structure into music does reflect a deeper reality. "When you examine a molecule in a textbook, it appears static," Buehler observes. "But it's not static at all. It's constantly moving and vibrating. Every bit of matter represents a set of vibrations. And we can employ this concept as a means of describing matter."
The technology does not yet allow for any kind of directed modifications—any changes in properties such as mechanical strength, elasticity, or chemical reactivity will be essentially random. "You still need to conduct the experiment," he notes. When a new protein variant is produced, "there's no way to predict its behavior or properties."
The team also created musical compositions developed from the sounds of amino acids, which define this innovative 20-tone musical scale. The artistic pieces they constructed consist entirely of the sounds generated from amino acids. "There are no synthetic or natural instruments used, demonstrating how this new source of sounds can serve as a creative platform," Buehler explains. Musical motifs derived from both naturally existing proteins and AI-generated proteins are utilized throughout the examples, and all the sounds, including some that resemble bass or snare drums, are also generated from amino acid vibrations.
The researchers have developed a free Android smartphone application, called Amino Acid Synthesizer, that enables users to play the sounds of amino acids and record protein sequences as musical compositions.
"Markus Buehler possesses an exceptionally creative mind, and his explorations into the inner workings of biomolecules are advancing our understanding of the mechanical response of biological materials in a profoundly significant manner," says Marc Meyers, a professor of materials science at the University of California at San Diego, who was not involved in this research.
Meyers adds, "The application of this creativity to music represents a novel and intriguing direction. This is experimental music at its finest. The rhythms of life, including the pulsations of our heart, were the initial sources of repetitive sounds that gave birth to the marvelous world of music. Markus has descended into the nanospace to extract the rhythms of amino acids, the fundamental building blocks of life."
"Protein sequences are complex, as are comparisons between protein sequences," notes Anthony Weiss, a professor of biochemistry and molecular biotechnology at the University of Sydney, Australia, who also was not connected to this research. The MIT team "provides an impressive, entertaining and unusual approach to accessing and interpreting this complexity. ... The approach benefits from our innate ability to hear complex musical patterns. Through harmony and discord, we now have an entertaining and useful tool to compare and contrast amino acid sequences."
The research team also included scientist Zhao Qin and Francisco Martin-Martinez at MIT. This groundbreaking work was supported by the U.S. Office of Naval Research and the National Institutes of Health.
