Research On Terahertz Time Domain Spectroscopy Analysis Methods For Cancer Markers Methylation Inspection | | Posted on:2022-07-04 | Degree:Master | Type:Thesis | | Country:China | Candidate:P Y Chen | Full Text:PDF | | GTID:2491306335466354 | Subject:Control Science and Engineering | | Abstract/Summary: | PDF Full Text Request | | Terahertz time domain spectroscopy(THz-TDS)technology,as a new nondestructive testing method,has been widely studied and discussed in biomedical field in recent years,especially in tumor detection and human diseases.Based on the National Natural Science Foundation of China project and the actual detection needs,this thesis studied the detection of methylation of potential human cancer markers using terahertz time domain spectroscopy.The detection and analysis methods of THz time-domain spectroscopy were studied for the methylation of small molecule DNA base and large molecule protein respectively.Taking cytosine and bovine serum albumin as examples,the terahertz spectral response mechanism of cytosine(normal)and 5-methylcytosine(methylated),qualitative detection of protein methylation and regression analysis of methylated protein concentration were studied.This thesis tried to solve the problem that it was difficult to extract the characteristic parameters of protein solution in terahertz spectrum.The difference between 5-methylcytosine and cytosine in terahertz time domain spectrum was analyzed from the mechanism and experiment,and the feasibility of terahertz time domain spectrum detection method for methylation detection was explored.It provided technical support for the application of terahertz time domain spectroscopy in clinical auxiliary diagnosis.The main work and innovation are as follows:(1)The THz response mechanism and analysis methods of 5-methylcytosine,a small molecule cancer marker and methylation product of cytosine,were studied.In order to better explain the feasibility of 5-methylcytosine detection by terahertz spectroscopy,according to density functional theory,a molecular simulation process was designed by optimizing the molecular structure and establishing the vibration mode.The modeling and analysis of cytosine and 5-methylcytosine were carried out by using quantum chemistry software including Gaussian and Materials Studio.The theoretical resonance peaks of the two materials in terahertz spectrum were obtained.At the same time,the THz time domain spectroscopy was used to detect cytosine and 5-methylcytosine,and the difference of THz spectral response between them was analyzed and compared.It was confirmed that there was a difference between the peak positions of cytosine and 5-methylcytosine in THz time domain spectrum from both theoretical and experimental perspectives,and THz technology could be used to distinguish the two substances.(2)The qualitative detection method of methylated protein solution by terahertz time domain spectroscopy was studied.Using bovine serum albumin(BSA)solution and methyl BSA solution as protein solution model,the difference of terahertz time domain spectral response of two proteins was studied,and the reason for the difference was explained by the hydration effect of protein.In order to solve the problems of high signal-to-noise ratio,high spectral similarity and insignificant terahertz response of solute in protein solution,a feature extraction method based on multi-resolution wavelet information entropy was proposed,which paid more attention to the details of time-domain spectrum and achieved the purpose of increasing the difference of terahertz spectral characteristics of different solutions.The effectiveness of multi-resolution wavelet information entropy in extracting terahertz spectral features of protein solution was verified by the case study of different concentrations of bovine serum albumin solution and methylated bovine serum albumin solution.At the same time,the absorption coefficient,refractive index and other parameters were used to distinguish the two solutions.The experimental results showed that the multi-resolution wavelet information entropy was more sensitive to the methylation of bovine serum albumin.It provided new terahertz spectral characteristic parameters for protein methylation detection and enhanced the accuracy of protein detection.(3)The concentration regression method of methylated protein solution based on terahertz time domain spectroscopy was studied.The degree of protein methylation is different in different diseases.THz-TDS was used to quantify the concentration of methylated bovine serum albumin solution.The terahertz absorption coefficient of protein solution changed nonlinearly with the change of solution concentration.The terahertz absorption coefficient data of the whole spectrum had strong correlation and redundancy.Based on the above problems,a feature extraction method of protein solution terahertz time domain spectrum based on the maximum information coefficient method was proposed,and the feature frequency points which were closely related to the concentration were selected.The concentration regression model of methylated bovine serum albumin solution was established based on absorption coefficient of the selected characteristic frequency point.Compared with the regression model based on full spectrum,the effectiveness and accuracy of the maximum informatioan coefficient in protein solution concentration detection were verified.It laid a technical foundation for the detection of protein methylation concentration in complex environment in the future.To sum up,terahertz time domain spectroscopy analysis methods for cancer markers methylation inspection were investigated.THz wave detection experiment,simulation and spectral analysis of 5-methylcytosine,a potential cancer marker,were carried out to verify the feasibility of detecting cytosine methylation by terahertz time domain spectroscopy.The terahertz time domain spectroscopy of methylated bovine serum albumin solution was analyzed,and the feature extraction method of terahertz time domain spectroscopy was proposed for category detection and concentration regression.The research could be used for reference in the application of terahertz time domain spectroscopy in the detection of biomolecule methylation. | | Keywords/Search Tags: | Terahertz time domain spectroscopy, Density functional theory, Multi-resolution wavelet information entropy, Cytosine, Methylation, Bovine serum albumin, Maximum information coefficient, Quantitative detection | PDF Full Text Request | Related items |
| |
|