Identification Of Environmental Microorganisms Based On Multi-wavelength Transmission Spectroscopy And Single-cell Optical Tweezers Raman Spectroscopy

In recent years,with the continuous development of human industrialization,ecological balance has been accelerated destruction,environmental pollution has become more and more serious;With the deterioration of people’s living environment,there are more and more kinds of microorganisms that are harmful to human beings,and they are more and more infectious and destructive.After the water is polluted,people may cause the outbreak of waterborne diseases such as typhoid fever,hepatitis and cholera by drinking or contacting the polluted water,which seriously threatens human health.While the traditional detection methods for microorganisms,such as morphological observation,nucleic acid probe,polymerase chain reaction,etc.,although these methods are accurate in the measurement results,but the operation is cumbersome,time-consuming,manpower and financial consumption.In recent years,optical technology has been widely used in the study of complex biological systems in the detection of microorganisms,which is favored by people as a kind of fast speed and non-damage.In this thesis,environmental microorganisms as the research object,the application of multi-wavelength transmission spectroscopy and single-cell optical tweezers Raman spectroscopy technology to study bacteria,and use different data analysis methods to analyze the obtained spectral data,the obtained results are mainly divided into four parts:(1)Use of existing strains:Micrococcus Aloe,Staphylococcus aureus,Bacillus Lysine Globulus,Bacillus thuringiensis(Bacillus),Bacillus Aloe,Bacillus Lysine Globulus,Bacillus thuringiensis(Bacillus),Bacillus Lysine Globulus,Bacillus thuringiensis(Bacillus)Thuringiensis)and Rheinheimera aquimaris were cultured to obtain the related strain samples,and the ultraviolet spectrum data of different strain samples were obtained by ultraviolet spectrophotometer.Spectral data were processed by summation normalization and max-min normalization respectively.Certain spectral data of each strain were selected to make the average spectrum of the strain,and the obtained average spectrum was regarded as the template spectrum of the strain.After the establishment of spectral template,the unknown samples of five species were identified by similarity calculation methods such as similarity theory.The overall recognition rates of five species were 90%,100%,90%,92%,and 100%,finally confirming the feasibility of the identification method.(2)Conduct field tests according to the method established in Part(1);The establishment of bacterial template in water body: Samples were collected from the sewage outlet near Yao Wharf on the Yongjiang River in Nanning.After purification and culture,16 S r DNA sequencing was performed on the bacterial samples to obtain the bacterial species of each sample.Then UV spectrophotometer was used to collect the spectral information of each sample,and a bacterial spectrum template was established.Bacteria identification:Samples were collected from sewage outfall near Yongjiang River Pump Station in Nanning.Bacteria samples were identified twice by using the established bacterial spectrum template and 16 S r DNA sequencing method.The differences between the newly established method and 16 S r DNA sequencing method were compared.It shows the practicability of the new method.(3)In the natural growth state of bacteria,the bacterial samples cultivated will have different growth stages,so as to ensure that the bacterial samples used for measurement are in the same growth stage,which can greatly reduce the experimental error;The synchronized cells of five bacteria were obtained by Percoll continuous density gradient separation method.Raman spectral curves of different bacteria were obtained based on Raman spectroscopy,and multivariate function analysis method combined with data preprocessing method was used to analyze the spectral data.The main differences of the components among different species of bacteria were analyzed by means of the load diagrams in the results of Discriminant Function Analysis(DFA).According to the results obtained,the main reason for distinguishing different bacterial species is the different contents of carbohydrate,protein and nucleic acid in bacteria.(4)Synchronous cells of each bacterium were obtained by the improved Percoll continuous density gradient separation method.After obtaining the Raman spectra of each bacterium,Principal Component Analysis(PCA)was performed among different bacteria.Based on PCA results and the Soft Independent Modeling of Class Analogy(SIMCA)method,modeling and analysis of the bacteria were carried out.The blind bacterial samples were selected and compared with the established template.The recognition rates of the five bacteria were as follows:The identification rate of Exiguobacterium,Acinetobacter towneri and Bacillus cereus was 100%.The identification rate of Acinetobacter baumannii and Aeromonas caviae was 94.4%,which finally confirmed the reliability of the established method.