Artificial intelligence injects wisdom and kinetic energy into scientific research (self-reliance and self-improvement of science and technology)


Artificial intelligence injects wisdom and kinetic energy into scientific research (self-reliance and self-improvement of science and technology)

  Core reading

  A few days ago, the Ministry of Science and Technology and other six departments issued a document, focusing on creating a number of major scenarios and expanding the application of artificial intelligence. High-level scientific research activities are one of them. Nowadays, China’s artificial intelligence technology is developing rapidly, and it has advantages in data acquisition, experimental prediction and result analysis. Life science, mathematics, chemistry, space science and other disciplines have embraced artificial intelligence. Rich application scenarios also feed back the development of technology and promote the upgrading of industrial wisdom.

  From daily life to scientific research, today, China’s artificial intelligence technology is developing rapidly, and data and computing resources are increasingly abundant. Application demand is an important driving force for technological progress, and new technologies are often constantly improved and matured in "use". In order to promote the landing of artificial intelligence, a few days ago, the Ministry of Science and Technology and other six departments jointly issued the "Guiding Opinions on Accelerating Scene Innovation and Promoting High-quality Economic Development with High-level Application of Artificial Intelligence", focusing on creating several major scenes and expanding the application of artificial intelligence. High-level scientific research activities are one of them.

  As a means of empowerment, how can artificial intelligence bring new research methods, and how can economic development inject "wisdom kinetic energy"?

  Close integration helps scientific research to be more efficient and accurate.

  Among many disciplines, life science research and artificial intelligence are closely integrated, and one of the hot directions is to predict protein structure.

  Protein has a three-dimensional structure, and its primary structure (sequence) is composed of several amino acids in series. The three-dimensional structure determines the function of protein in cells, and many diseases are caused by the important protein structural abnormality in the body. Therefore, only by drawing a "three-dimensional map" of important protein in human body can we find the target of drugs acting on human body, thus developing accurate and effective new drugs.

  Traditionally, scientists have observed the three-dimensional structure of protein by means of cryoelectron microscopy, X-ray and nuclear magnetic resonance, but this process is time-consuming, laborious and expensive. "Taking the cryomicroscope as an example, it costs tens of millions of yuan to arrange an observation platform, and it will take a long time for researchers to draw the protein structure." He Jingzhou, head of Baidu’s propeller bio-computing platform, said.

  Due to the high difficulty, long experimental period and high cost, the number of three-dimensional structures of protein observed by traditional methods is very limited. In contrast, amino acid sequencing is much easier. Why can’t we predict the structure of protein based on the amino acid sequence? As early as 1972, American biochemist Christian anfinsen put forward this idea in his Nobel Prize acceptance speech.

  It is precisely artificial intelligence that accurately predicts the three-dimensional structure of proteins from their primary structure. However, human beings have made slow progress in trying to analyze protein Group. It is explained that, on the one hand, because the existing biological data is small in quantity and low in quality, deep learning lacks enough samples; On the other hand, because the artificial intelligence algorithm is mature, it also needs a process.

  In recent years, with the sharp increase of biological data and the optimization of artificial intelligence technology, scientists have established more accurate prediction models. In December 2020, in a competition, the artificial intelligence program "Alpha Folding" shined brilliantly, and its predicted results were similar to most experimental data. This proves that artificial intelligence has been quite accurate in predicting the structure of protein.

  Nowadays, with the help of artificial intelligence, the work that may have taken several years can now be completed in a few minutes, and some protein structures that cannot be observed by traditional methods can also be analyzed.

  It is understood that using artificial intelligence, researchers have predicted the structure of more than 200 million protein species of about 1 million species, covering almost every protein that has been recorded by the scientific community. This will have a great impact on the field of structural biology, which may lead to a paradigm change in life science research and enhance human understanding of life.

  It has broad prospects and plays a role in many fields such as life science.

  Artificial intelligence has entered the field of vision of life science research, and the demand of biomedical industry is an important driving force. According to reports, in the biopharmaceutical industry, the types of drugs that can be developed for every $1 billion invested have been declining. The research and development of new drugs is becoming more and more difficult, and the cycle is getting longer and longer. There is an urgent need for new methods to break through, and artificial intelligence has high hopes.

  Not only does it accelerate the research and development of new drugs, but artificial intelligence is playing an important role in more and more fields of life sciences.

  At the beginning of this year, a protein prediction model run by Chengdu Center of National Supercomputing helped the team of Wheat Research Institute of Agricultural College of Sichuan Agricultural University successfully analyze the molecular mechanism of PGS1 regulating seed development and affecting yield, and provided theoretical basis for cultivating high-yield and high-quality wheat materials. Researchers say that without artificial intelligence, it is difficult to make this breakthrough efficiently.

  Researchers also try to introduce artificial intelligence technology into vaccine design. For example, compared with protein vaccine and DNA vaccine, mRNA vaccine has the advantages of fast mass production and good anti-infectivity, but its stability and immunogenicity are relatively poor. To make up for these shortcomings, researchers have been hoping to optimize the sequence design of mRNA vaccine to make it more stable and more immunogenic. With higher efficiency and lower cost, the intervention of artificial intelligence is expected to provide new ideas for vaccine research and development.

  Precision therapy is also a stage for the application of artificial intelligence. Through the method of machine learning, in theory, artificial intelligence can decode the human immune system and explore the complex immune laws of some diseases more accurately, thus helping people understand diseases and develop therapeutic drugs and methods more efficiently and pertinently.

  With the increase of human data and knowledge accumulated in new drug research and development brought by genomics research, and the continuous iteration of machine learning algorithms, industry experts believe that artificial intelligence has broad prospects in the field of life sciences. Some researchers even imagine to build a unified knowledge map by relying on a powerful biological computing engine and using a large amount of biological data to promote the understanding of life and health.

  Rich scenes, pushing the application to a higher level.

  Walking into life science research, artificial intelligence brings new methods, and it is also expected to open a new era of biomedical industry.

  Experts in the industry said that seizing the new opportunities of intelligent drug design and strengthening the layout of artificial intelligence and biomedicine will help us to take the lead in the new track of new drug research and development.

  Although artificial intelligence+biomedicine has developed rapidly, it has just started on the whole. He Jingzhou believes that the biomedical industry has a history of hundreds of years, with mature and perfect research processes, industrial chains and division of labor. The improvement of artificial intelligence is only some of them. "Biomedical is related to life and health, and the development of the industry should be steadily advanced, maintaining rationality and awe." Up to now, there is no new drug discovered entirely by artificial intelligence in the world. Some products developed by artificial intelligence are still a long way from the real market.

  Seeing the advantages of artificial intelligence technology in data acquisition, experimental prediction and result analysis, mathematics, chemistry, materials science and space science have also embraced artificial intelligence.

  The performance of lithium batteries varies with different material compositions. In response to the demand for lithium batteries in a variety of scenarios, researchers hope to design a suitable lithium battery system by optimizing the material combination.

  "In the past, the design of material systems mainly relied on manual experiments, which was very inefficient." Zhang Qiang, a professor at the Department of Chemical Engineering in Tsinghua University, said that at present, he is leading the team to predict molecular properties with artificial intelligence, so as to find energy materials more efficiently and accurately and design more valuable and safer battery systems.

  According to estimates, the scale of China’s artificial intelligence core industry exceeds 400 billion yuan, and the number of enterprises exceeds 3,000. Thanks to the strong demand brought by massive data processing and the experimental soil provided by rich application scenarios, China is at the forefront of the world in the fields of computer vision and speech recognition.

  Experts in the industry suggest that it is necessary to give full play to the advantages of rich application scenarios in China to promote the application of artificial intelligence to a higher level. Creating a major scene around high-level scientific research activities will promote the application of artificial intelligence in China and inject "wisdom and kinetic energy" into high-quality economic development.


admin administrator