IT House news on February 10, according to the official website of the University of Science and Technology of China, the team of Professor Liu Haiyan and Associate Professor Chen Quan of the University of Science and Technology of China adopted a data-driven strategy to open up a new protein design route from scratch.
IT House learned that the relevant results were published in Nature on February 10, Beijing time, under the title of "Backbone Neural Network Energy Function for Protein Design".
According to reports, protein is the basis of life and the main executor of life functions, and its structure and function are determined by amino acid sequences. At present, proteins that can form a stable three-dimensional structure are almost all natural proteins, and their amino acid sequences are naturally evolved over a long period of time. When the structure and function of a natural protein cannot meet the needs of industrial or medical applications, the structure and sequence of a specific functional protein need to be designed.
The relevant team of the University of Science and Technology of China has long been deeply engaged in basic research and applied basic research in the direction of computational structural biology. The team of Professor Haiyan Liu and Associate Professor Quan Chen has been committed to developing data-driven protein design methods for more than ten years, establishing and experimentally verifying the ABACUS model of amino acid sequence design for a given backbone structure, and then developing an SCUBA model that can design a new backbone structure from scratch when the amino acid sequence is pending (Figure 1). SCUBA adopts a new statistical learning strategy, based on kernel density estimation (or neighbor count, NC) and neural network fitting (NN) method, from the original structural data to obtain the analytic energy function in the form of a neural network, which can reflect the high-dimensional correlation between different structural variables in the actual protein structure in high fidelity, continuously and extensively search the main chain structure space under the premise of uncertain sequences, and automatically generate a "highly designable" backbone.
Figure 1.Principles of protein design using the SCUBA model. (a) The minima of the energy plane of the SCUBA backbone corresponds to the designable backbone structure of the protein, i.e., the lowest free energy structure under a particular amino acid sequence; (b) the statistical energy term expressed in neural networks in SCUBA; and (c) and (d) the method framework for learning the analytical energy function from the raw data of the protein structure using the near neighbor counting (NC)-neural network (NN) method | Source: The official website of the University of Science and Technology of China
Theoretical calculations and experiments have proved that the design of the backbone structure with SCUBA can break through the limitation that only natural fragments can be used to stitch together to produce a new backbone structure, significantly expand the structural diversity of de novo design proteins, and then design a novel structure different from known natural proteins. The SCUBA Model + ABACUS Model forms a complete toolchain of artificial proteins capable of de novo designing with entirely new structures and sequences, and is currently the only experimentally validated protein de novo design method outside of RosettaDesign that complements and complements them. In the paper, the team reports on the high-resolution crystal structures of nine de novo designed protein molecules (Figure 2) whose actual structures are consistent with the design model, where 5 proteins have novel topologies that have not yet been observed in native proteins.
Figure 2.Comparison of the high-resolution crystal structure (sky blue) of the de novo design protein with the design model (green) | Source: The official website of the University of Science and Technology of China
Original link:
https://www.nature.com/ articles / s41586-021-04383-5