Study On Identification Methods In The Detection Of Transgenic Material Based On Terahertz Time Domain Spectroscopy

With the development of genetic engineering technology, Humans have the ability to be exogenous gene into an organism, it makes the biological have other new features, for example:Insect resistance, stress resistance etc.. These transgenic species will pose a potential threat to human health, ecological environment and ethics and so on. Therefore, it is very important to use research on the detection and identification of the genetically modified organisms. At present, there are some kinds of transgenic detection method mainly including protein-based and acid-based. Although these methods can effectively identify genetically modified organisms, these methods also exist some drawbacks, which include the detection of protein-based are only suitable for inspecting the raw product and the acid-based have the high cost in the process of testing. Terahertz (THz) technology is a very hot topic in the field of the frontier, because of the unique nature of THz wave(such as perspective,strong security and good spectral resolution), it makes the THz has a very broad application prospects in the aspects of physical test.This research mainly studies the detection of genetically modified organisms (including transgenic cotton, transgenic beet and its parents) by using the spectroscopy analysis and optical chemical measurement technology. The purpose of this paper is to demonstrate the feasibility analysis technique for detection of genetically modified organisms by using THz spectroscopy, and establish rapid a rapid detection system of genetically modified organisms based on THz spectroscopy technology, it can provide the basis for transgenic detection instrument of Independent intellectual property rights.The main research contents of this paper are as follows:(1) Using terahertz time-domain spectroscopy to identification of transgenic cotton was analyzed. An approach to support vector machine model based on adaptive affinity propagation clustering algorithm is proposed, and it is successful to apply on THz spectroscopy classification of transgenic cotton. On the basis of studying the absorption curve of several transgenic cottons, the THz spectroscopy information is extracted are extracted in the range of 0.2THz to 2.2THz as the characteristic parameters of identification. Firstly, in the iterative cycle continuously, clustering and labeling to the unlabeled training samples by using adaptive affinity propagation clustering algorithm, and update the training data of the support vector machine until the all unlabeled sample is labeled. Secondly, after the model is established can be used to test the sample. Because it can automatically classification and labeling of samples, and does not need the artificial marking model of training samples, so, it can reduce the error bring from artificial labeled. The research results show that the support vector machine model based on adaptive affinity propagation clustering algorithm has high recognition rate for transgenic cotton, and it can provide an effective and nondestructive method in identification of genetically modified organisms by using THz spectroscopy.(2) Methods for identification of transgenic substance based on THz-TDS have been investigated. To avoid the drawbacks from the traditional pattern recognition, this paper proposes a novel identification methods based on Hyper Sausage Neuron(HSN) combined with principal component analysis (PCA). Using transgenic substance to verify the feasibility and effectiveness of this model. First of all, Principal component analysis (PCA) is applied to extract the features spectrum of transgenic substance. In the next place,using the features spectrum instead of the original spectrum data to fed into the HSN pattern recognition for test samples. The experimental result shows that the HSN model not only can correctly classify different types of transgenic substance, but also can reject identity non similar samples in the same type. The proposed approach provides a new effective method for detection and identification of GMOs by using THz spectroscopy.(3) This section reports the first use terahertz time-domain spectroscopy combined with metamaterials to identify genetically modified organisms. A split ring resonator (SRR) structure was prepared which can provide negative magnetic response based on the complex boundary condition of metamaterials. Because of the different liquid have different dielectric constant,when the liquid membrane is added to the split ring resonators void, the dielectric constant around split ring resonators is changed, the split resonators ring terahertz response peak is moved, and the redshift phenomenon is produced. The different transgenic products will be identified by using this redshift phenomenon. The most feature of this method is not needed to know the samples complex spectral information. The experimental results show that the transgenic and non-transgenic products with a similar character can be distinguished clearly by compares split resonators ring terahertz response peaks.The results suggest that the proposed method may be employed to sense applications.