Author: Chen Yi
From gene fragments, DNA molecules, gene regulatory networks and signaling pathways to the artificial design and synthesis of cells, synthetic biology is developing rapidly. Focusing on the essential issues of cell structure and function, as well as cutting-edge technologies such as cell engineering and artificial cells, the sub-forum "Exploring Synthetic Cells" of the 5th International Joint Laboratory Forum of the World's Top Scientists was held in Shanghai Science Hall this afternoon, and the current progress and future development of the field were hotly discussed.
James Rothman, 2013 Nobel Prize in Physiology or Medicine, Head of the Department of Cell Biology at Yale University, Irwin Nel, 1991 Nobel Prize Laureate in Physiology or Medicine, Director and Scientific Member of the Max Planck Institute for Biophysical Chemistry, Stefan W. Hull, 2014 Nobel Prize in Chemistry and Director of the Max Planck Institute for Multidisciplinary Sciences, 2012 Lasker Prize for Basic Medical Research, Deputy Director of the Howard Hughes Institute for Medical Research, Ronald Weil, Executive Director of the Janelia Research Park, Xu Jinbo, winner of the Sloan Research Award and tenured professor at the Toyota Institute of Computing Technology in Chicago, Yuan Yingjin, academician of the Chinese Academy of Sciences and vice president of Tianjin University, Zeng Anping, academician of the German Academy of Engineering and founding director of the Center for Synthetic Biology and Biointelligence at Westlake University, and Shao Yangyang, researcher of the Hundred Talents Program of the College of Life Sciences, Zhejiang University, and other scientists gave reports respectively. The forum was chaired by Fan Chunhai, academician of the Chinese Academy of Sciences and dean of the School of Chemistry and Chemical Engineering, Shanghai Jiao Tong University.
Artificial intelligence in predicting and designing proteins
What can and cannot be done?
Recently returned structural biologist Yan Ning and her team are known for their analytical proteins. The speed of this work is now being updated by Alpha Fold2, Google's second-generation deep learning algorithm from DeepMind. Using computational methods to study proteins, especially the use of AI technology to predict the three-dimensional structure of proteins and design some proteins that do not exist in nature, is also Xu Jinbo's work for a long time. He introduced that in the past 30 years, artificial intelligence has done a lot of good work in predicting protein structure, changing the way many microbiologists work. For example, in 2020, DeepMind developed Alpha Fold2 based on the first generation of deep learning algorithm developed by Xu Jinbo in 2016 to predict the three-dimensional structure of proteins, and successfully predicted the three-dimensional structure of proteins according to gene sequences. Protein design can be used to prevent, diagnose and treat disease, as well as in industry, agriculture and environmental protection. In addition, using AI to create or optimize some enzymes is also very promising. However, because the sequencing-based method is a very large project, there are still many problems that AlphaFold2 cannot solve. For example: predict how multiple proteins or antigens or antibodies bind; Designing proteins or optimizing the function of proteins produces variations to achieve the desired function; When the protein is large and has multiple domains, predict the orientation of all domains; Predict the structure of orphan proteins without "siblings", etc.
Designing proteins in different situations faces different problems, such as some to optimize amino acid sequences to obtain specific protein shapes, and some hope that the designed proteins bind to different targeted receptors, so different artificial intelligence algorithms are required to design different proteins. It is reported that Xu Jinbo's laboratory is currently studying antibodies, small proteins and biosensors for treatment, and has been able to use artificial intelligence to design protein backbones containing more than 300 amino acids, whereas in the past, the scientific community could only design some small proteins.
Proteins do not function alone in cells, but in most cases they function by interacting with other protein molecules or with other macromolecules such as DNA and RNA. This means that to really understand the function of proteins, you also need to understand how individual proteins bind to other molecules. For example, how two proteins are functioned when they are combined; How proteins work when they are combined with DNA and RNA; How antibodies and antigens interact. The CDR region of the antibody protein is homogeneous with the AI algorithm, which improves the affinity between the antibody and the antigen.
Artificially designed proteins are more stable than proteins that already exist in nature, and can perform a certain function in extreme situations such as the absence of energy or high temperatures. In addition, artificial intelligence can also be used to design amino acid sequences to match the backbone and improve the stability of specific proteins such as enzymes and antibodies. Xu Jinbo said that artificial intelligence technology has played a very important role in the design of proteins of different sizes and conformations, and can also help design more and more useful proteins in the future, including new biomaterials that can be used to reduce pollution and improve the environment. His team is now well able to model more than 50 percent of protein sequences in nature.
In the past, in order to train AI models, a lot of laboratory data was usually required, Xu Jinbo said that the training model designed according to his team's new algorithm can already obtain good results without the need for a large amount of data, and less data can obtain good results. In the past, antibodies worked by binding to specific target proteins to fight diseases such as infection or cancer, but because antibodies are large proteins, difficult and expensive to manufacture, and lack stability, a new method of producing protein drugs led by David Baker, director of the Institute for Protein Design at the University of Washington, uses computer design to target small molecule-binding proteins that target important proteins in the body, such as insulin receptors and surface proteins of viruses. This ability to produce new proteins that bind tightly to any desired target protein is a paradigm shift in drug development and the broader field of molecular biology. This progress may help develop new treatments for many diseases, including cancer, diabetes, and the new crown.
Synthetic biology and biomanufacturing
Global challenges can be addressed in many areas
Make a difference
Owen Nel said the scientific community has a task of using cell signaling networks to create life. In the 1970s, people only knew about 5-10 different types of ion channels, through technological improvements and channel research, now scientists have found that there are ion channels in many types of cells, they play different roles, through voltage, neurotransmitters, stress, odor, hot and cold, etc., may activate ion channels.
Zeng Pingan pointed out that in the next 50 years, the world will face ten major challenges such as energy, water, food, environment and climate, poverty, terrorism and war, disease, education, democracy, and population, and synthetic biology and biomanufacturing are very important to meet these challenges in many fields. There is a great need for efficient cell factories and synthetic cells. Biomanufacturing using sugar as a raw material today is unsustainable, and in the future it may need to replace sugar with carbon atoms in carbon dioxide. At present, the use of microorganisms and plants to fix carbon dioxide is not very efficient, and new paths need to be explored in this regard.
Shao Yangyang proposed that if some corrective genes can be cloned, imported into specific cells, and then accurately knocked out in the genome, it may be expected to treat some not uncommon chromosomal diseases.
Although the theme of the forum was "Exploring Synthetic Cells", James Rothman, winner of the 2013 Nobel Prize in Physiology or Medicine, "relentlessly" poured a little "cold water" on his peers. He believes that people's current understanding of the concepts of "synthetic biology" and "synthetic cells" may still be somewhat confusing, and synthetic cells may not be realized in the near future.