Experimental Research On ML Recognition Of Mixed Amino Acids Based On Chemical Chip Technology

With the rapid development of economy and society,people’s demand for material life is constantly increasing.Amino acids,as the basic components of all life,have an extremely important impact on the natural environment during the decomposition of substances.Therefore,amino acids naturally become the focus of people’s attention,and the environmental pollution caused by the business and industry related to amino acids will also occur.In order to better understand the environmental behavior of amino acids in high protein decomposition or healthy human life,it is necessary to timely analyze and study amino acids.In terms of complex chemical systems,the complexity of the material composition will often lead to some linear independent reactions or nonlinear coupling reactions between various substances within the system.Therefore,the identification and analysis of mixed amino acids are often very difficult due to the interactions between their internal substances.In traditional scientific research,large instruments are often used to detect the composition of mixed amino acids,or thousands of experiments are conducted to look for the interaction between three or four amino acids for relevant analysis and inference.However,traditional instruments can only detect the existence and content of a certain kind of substance or pollutant,but cannot deeply understand its environmental effects.At the same time,the traditional research methods are tedious,long experimental cycle,expensive experimental costs and other shortcomings.In the face of the shortcomings of the traditional technology in the study of complex chemical systems,the research group through several years of efforts to explore a technical means.This technique can not only record and present the internal information of the mixed amino acids,but also predict the mixed amino acids through the correlation model.Based on the dynamic equilibrium relationship between substances,we propose a strategy of recording and presenting the internal information of amino acids in complex chemical systems through chemical interferents and chemical chromogenic agents.Under the guidance of high-throughput technology,we used inkjet printing technology to combine thousands of organic combinations of chemical interferents and chemical colorants on the corresponding printing carrier.So that they are distributed independently on the printed carrier according to a certain rule,and finally form a chemical chip that can produce an order of magnitude of micro-reaction.Similarly,we can use inkjet printing to print amino acids in different amounts on hundreds of chemical chips at the same time.At that time,the internal information of amino acids in the complex chemical system will be recorded and presented by the chemical chip in the form of color change,forming the image of the reaction chip.Through the preliminary digital analysis of’the reaction chip image,we have a relatively simple understanding of the internal information of the mixed amino acids.After that,we use ML to build the relevant model for its internal information.This model can be used for the identification and analysis of amino acids in complex chemical systems and provides some reference for the study of mixed amino acids.Main conclusions:(1)To construct a method for recording and presenting the internal information of mixed amino acids in the natural environment.Chemical reagents were screened out based on the relationship between the internal dynamic balance of substances,and then chemical chips were formed by accurately distributing chemical reagents on the printing carrier using inkjet printing technology under the guidance of high-throughput technology.The chemical chip can record and express the internal information of the mixed amino acid in the form of color change..(2)Digital processing of image of reaction chip.The RGB value or gray value of the image of the reaction chip is extracted by relevant software,and the information contained in the image of the reaction chip is presented more intuitively in the form of contour map,RGB histogram and 3d curves diagram,so that we can compare and analyze it.(3)Preliminary study on the internal information of mixed amino acids.After the digital processing of the image of the reaction chip,a series of comparative analysis was conducted on the corresponding digital results under various experimental conditions,and the internal information of the mixed amino acids was preliminarily understood.(4)Research on the performance of chemical chips.The error analysis of the reaction chip indicates that the chemical chip prepared in this study has better accuracy.At the same time,we put forward some Suggestions for its application in other fields in the discussion of the applicability of chemical chips.(5)The identification and analysis model of mixed amino acids was established.Through the correspondence between the image of the reaction chip and the quantity and type of the reaction sample,the original network was trained and learned deeply by using ML.Finally,a network model of internal information of mixed amino acids is established,which provides support for the identification and analysis of mixed amino acids.