In keeping with researchers at Massachusetts Institute of Technology (MIT) within the US, they both gain from being composed of structures over a variety of scales be that notes, chords, and melodies or amino acids, proteins and collagen matrices. Taking the analogy further Buehler and colleagues have translated the vibrations of amino acids and the longer-range buildings of the proteins they type right into a musical framework. AI algorithms trained on this musically transcribed protein data may devise recent amino acid music based on musical principals learned from the coaching data set, which the researchers then translate again into protein structures.
“What we’ve been attempting to do in a variety of different ways is to find methods of predicting the functionality of a protein from its sequence, and that is a very tough thing to do,” Buehler tells Physics World. For several years Buehler and his group at MIT have studied supplies including spiders’ webs and nacre to recognize the hierarchical structures behind their spectacular mechanical properties.
He defines that current approaches to relating protein structure and performance generally depend on long computational resource-hungry molecular dynamics simulations to solve equations approximating the quantum mechanical interactions on the molecular scale to decide how the protein folds, and how it operates. “one of the directions we have pursued is to suppose how we look at materials, and we realized that after we look at materials on the molecular scale, the atoms and molecules constantly vibrate. So we thought that maybe there is a way of capturing the spectrum of vibrations on the nanoscale and developing a model from that.”