AI-Chemical Artificial intelligence is liberating chemists' hands

Recently, Xiangshan Science Conference focused on green ecology and chemical industry, in addition to a variety of cutting-edge chemical and chemical technology, the application of artificial intelligence in this field has become a new topic of concern.
variables in the "synthetic chemistry process" does make it more difficult to prepare the amount and filter the conditions of the trout reaction. But in the age of big data and artificial intelligence, this will dramatically change

. Ma Dawei, deputy director of the Shanghai Institute of Organic Chemistry of the Chinese Academy of Sciences, told the conference that artificial intelligence is expected to help synthetic chemistry research to a new level.
"labour-intensive" will become the past
" in the field of organic synthesis, synthetic chemists in the last century claimed to give enough student and financial support to synthesize any complex small organic molecules. But to this day, whether it's new drug development or new material development, it's been found that the decisive step in the process is often compound synthesis. "Ma Da for the introduction.
explains that this is because, although most of the evaluation of the properties of compounds has entered the vodium screening phase, chemical synthesis remains in the manual operation phase.
in the field of chemical synthesis, because of the infinite changes under different chemical reaction conditions, the researchers designed the chemical synthesis route largely by hand - especially in the field of pharmaceutical molecular synthesis. This makes chemical synthesis more like a "labour-intensive" job.
in the age of artificial intelligence, this will gradually become a past. Computer programs based on big data and artificial intelligence are becoming more sophisticated and practical in assisting researchers in the design of chemical synthesis routes.
in April last year, a German research team published a paper in the journal Nature that said they could use artificial intelligence systems to perform chemical synthesis analyses at an unprecedented rate, which would greatly improve the efficiency of researchers in developing new drugs and other compounds.
the paper, the researchers used artificial intelligence systems to design a molecular synthesis route in just five seconds. Ma Da lamented that if the artificial design, 5 seconds are too late to see the molecular structure clearly, 5 minutes are not necessarily able to design the molecules.
Chenghui, an associate professor at Nanjing University's School of Chemistry and Chemical Engineering, gained a lot of inspiration when he learned that artificial intelligence algorithms recommend the synthesis route of molecules. He is also currently considering how artificial intelligence could help him with his chemistry research.
Li Chenghui recently discovered a new molecular in-loop reaction, hoping to find out if it also exists in other molecules. According to the previous research method, his workload will be very large. Because to detect whether this reaction is universal, a large number of experiments with different molecules are required. "With the help of artificial intelligence, you can do it in a targeted way." Lee Seung-hui said.
began to act as a right-hand man
" small-volume reaction condition screening equipment has been in some companies and universities into conventional weapons, large-scale reaction matrix robots of large-volume has been born. "Ma Dawei believes that the popularity of these devices has the potential to make a huge difference in future synthetic chemistry research, largely freeing synthetic chemists' hands and allowing researchers to spend more time on top-level designs for data analysis and reactions.
ma Dawei in an interview with Science and Technology Daily, the artificial intelligence program used to predict the normal chemical small molecule synthesis route has matured. Even if some complex organic small molecules such as pharmaceutical molecules, because the structural units are relatively single, synthesis reactions are not many, artificial intelligence programs can directly assist in the design of synthesis routes. This can be of great help to junior researchers.
"Now doing drug synthesis research, if the structure of a drug molecule is known, the AI system can analyze a number of synthesis routes based on its structure and recommend an optimal route." Li Chenghui said that if it had been before, such research would have required very experienced organic chemistry experts, but in the future, with the help of artificial intelligence, researchers working in inorganic chemistry could do the same.
when synthesized into more complex target molecules, such as complex molecules that take more than 20 steps to synthesize into the natural world, artificial intelligence programs can predict many different paths.
researchers may only be able to come up with more than 20 synthesis routes, and these programs may provide hundreds of possible synthesis routes, which is very helpful for researchers because some routes may never have been thought of. Ma dawei said.
Ma Dawei told Technology Daily that based on the predictive routes provided by artificial intelligence programs, experienced researchers can tell which routes are good and which are suggestive, which in turn helps researchers think about some questions. This provides more opportunities for chemical synthesis. Because, according to the hints of artificial intelligence programs, coupled with further deep thinking in the human brain, a better chemical synthesis route may be designed.
the future only need to
the prescription of the drug", Ma Dawei believes that the above changes for chemical synthesis, the significance is very significant. Because artificial intelligence programs may be like "fool machines" in the future, researchers will only need to follow its recommendations for chemical synthesis.
the past to see if a chemical reaction condition is feasible is the process of constant trial and error. It needs to be tested manually one by one and contains a lot of repetitive labor. Now, artificial intelligence can be used to quickly detect chemical reaction data, and then on the machine to test the response conditions of the trout, the entire system can do thousands of reaction conditions a day testing. Instead of labor, it's good to do 20 chemical reaction tests a day in a lab.
pharmaceutical companies have become the first users of such technologies by using artificial intelligence to aid in the design of chemical synthesis routes, Ma Dawei, a researcher, told Technology Daily. In addition to drug development, artificial intelligence can be used to recommend chemical synthesis routes in areas including compound synthesis, such as new material development.
"In the future, synthesize molecules that didn't exist in the past is like building a house, allowing artificial intelligence to map out the blueprint first, and researchers will just have to follow the blueprint to build it," he said. Ma dawei said.
Li Chenghui also believes that artificial intelligence in the field of chemistry will have great potential. In his view, the future application of artificial intelligence in the field of chemistry will be like artificial intelligence doctors, it can master and digest mass synthesis methods, synthesis routes, material structure and performance, and so on, researchers do chemical synthesis research, it can help analyze and solve many problems.
, the role of artificial intelligence is still complementary. It can free chemical researchers from heavy manual labor and provide some reference and reference for their research. But its recommended results also require researchers to use their expertise and experience to determine what is really feasible.
addition, the recommended chemical synthesis route for artificial intelligence is still in the model machine stage. Ma believes that chemical researchers need to constantly provide artificial intelligence with innovative strategies to promote the optimization of artificial intelligence systems, which in turn drive chemical researchers to conduct more in-depth research.
Link
Artificial intelligence robots help to quickly discover new molecules
As early as last year, chemists at the University of Glasgow in the UK discussed how to train an ARTIFICIAL organic chemical synthesis robot to automatically explore a large number of chemical reactions.
This "self-driven" system is based on machine learning algorithms that can discover new reactions and molecules, allowing for digital chemical data-driven methods to locate new molecules of interest, rather than being limited to a known database and conventional organic synthesis rules. The results may reduce the cost of discovering new drug molecules, new chemicals (including materials), polymers, and molecules for high-tech applications such as imaging.
team demonstrated the system's potential by simulating about 1,000 reactions using a combination of 18 different starting chemicals. After exploring about 100 possible reactions, the robot was able to predict with more than 80% accuracy which combinations of initial chemicals should be explored to produce new reactions and molecules. By exploring these reactions, they discovered a series of previously unknown new molecules and reactions. The researchers found four new reactions, one of which was classified as the first 1 percent of the most unique known responses.
researchers say this approach is a key step in digitizing chemistry, which will allow real-time retrieval of chemical space to help discover new drugs and cut costs, save time, improve safety, reduce waste, and help chemistry enter a new digital age.